Bio-based composite material twin-screw mixing granulation method

Abstract

The invention provides a double-screw mixing granulation method for a bio-based composite material, and belongs to the technical field of bio-based composite material production. The moisture content and the particle size distribution of wood flour are controlled through raw material pretreatment, and gradient temperature control mixing is adopted to establish a segmented heating system; wood powder thermal degradation parameters and melt flow resistance parameters are monitored in real time, a temperature control flow collaborative balance algorithm is started to dynamically adjust the temperature gradient and the screw rotating speed, a collaborative optimization framework of thermal stability control and fluidity improvement is constructed based on a double game model, and optimal processing parameters are obtained through iterative solution by using an improved alternating grey wolf algorithm. And adjusting the operation condition of the double-screw extruder according to the optimization result, extruding and molding through a wide-runner die head, and granulating and packaging. The technical problem that the processing parameters are difficult to collaboratively optimize due to mutual restriction of the thermal stability and the fluidity in the high-filling-amount processing process of the wood powder plastic composite material is solved.

Description

Technical Field

[0001] This invention belongs to the field of bio-based composite material production technology, and more specifically, relates to a method for twin-screw compounding and granulation of bio-based composite materials. Background Technology

[0002] Wood flour-plastic composites, as an environmentally friendly material, are widely used in building materials, furniture, and other fields. Traditional twin-screw extrusion processes achieve melt mixing of wood flour and polymer matrices by setting a fixed temperature gradient and screw speed. However, when the wood flour content reaches 65% or more, increasing the processing temperature, while reducing melt viscosity and improving flowability, exacerbates the thermal degradation of wood flour, leading to carbonization and discoloration of the composite material. Conversely, lowering the temperature to protect the thermal stability of the wood flour causes a sharp increase in melt viscosity, resulting in excessive extrusion pressure and torque overload. Existing processes typically employ empirical adjustment methods to find a balance between temperature and speed under fixed operating conditions, or improve flowability by increasing the amount of lubricant. These methods lack a real-time response mechanism to the dynamic changes in thermal degradation and flow resistance, and cannot establish a synergistic control relationship between multiple parameters. In other words, existing technologies present a technical problem where the thermal stability and flowability of wood flour-plastic composites are mutually constrained during high-filling-content processing, making it difficult to synergistically optimize processing parameters. Summary of the Invention

[0003] In view of this, the present invention provides a method for twin-screw compounding and granulation of bio-based composite materials, which can solve the technical problem in the prior art where the thermal stability and flowability of wood flour plastic composite materials are mutually constrained during high-filling-content processing, making it difficult to coordinately optimize processing parameters.

[0004] This invention is implemented as follows: A method for twin-screw compounding and granulation of bio-based composite materials includes the following steps: raw material pretreatment stage, gradient temperature-controlled compounding stage, real-time monitoring and data collection of wood flour thermal degradation parameters and melt flow resistance parameters, optimization of processing parameters based on a dual game model, and adjustment of the twin-screw extruder operating conditions according to the optimized parameters. Next, the extrudate is formed through a wide-channel die and then cooled by a water-cooling system before being pelletized and packaged. A first sub-game model for thermal stability control and a second sub-game model for improving flowability are constructed. These two sub-game models achieve global coordination through coupling terms such as wood flour thermal stability margin and screw conveying capacity. An improved alternating gray wolf algorithm is used iteratively to obtain the optimal screw speed matching value and the optimal feeding speed matching value. When the material temperature distribution shows an upward trend, the extrusion pressure shows a downward trend, and the residence time distribution remains relatively stable, a temperature-controlled flow synergistic balance algorithm is activated to dynamically adjust the temperature gradient and screw speed.

[0005] Specifically, the raw material pretreatment stage involves sieving the wood flour to a particle size range of 100-200 mesh, and then placing it in a drying device to dry it at a temperature of 80-90°C for 4-6 hours to reduce the moisture content to below 2%.

[0006] The raw material pretreatment stage further includes screening polyethylene resin according to the melt flow rate standard of 5g / 10min to 15g / 10min and mixing it with zinc stearate external lubricant at a mass ratio of 100:2 to 100:4 for later use.

[0007] Specifically, in the gradient temperature-controlled mixing stage, the treated wood flour and polyethylene resin are fed into a twin-screw extruder at a mass ratio of 65:35 to 75:25. The feeding section temperature is set to 100℃ to 120℃, the mixing section temperature to 140℃ to 160℃, the die head section temperature to 150℃ to 170℃, the screw speed to 200 r / min to 300 r / min, and the length-to-diameter ratio to 40 to 48.

[0008] The wood flour thermal degradation parameters include material temperature distribution and residence time distribution, and the melt flow resistance parameters include extrusion pressure and torque load.

[0009] The material temperature distribution value is obtained in real time by installing temperature sensors in the feeding section, mixing section and die head section of the twin-screw extruder, with a sampling frequency of 1Hz to 5Hz.

[0010] The residence time distribution value is calculated by adding a tracer to the material and detecting the tracer concentration change curve at the extruder outlet. The tracer is an inorganic pigment with good thermal stability.

[0011] The temperature-controlled flow collaborative balance algorithm achieves processing stability control by constructing a dynamic response relationship between material temperature distribution value and extrusion pressure value. When the material temperature distribution value is detected to be continuously rising, the temperature-controlled flow collaborative balance algorithm automatically reduces the temperature gradient amplitude between the feeding section and the mixing section.

[0012] In the temperature-controlled flow collaborative balance algorithm, when the extrusion pressure value decreases, the temperature-controlled flow collaborative balance algorithm simultaneously increases the screw speed to compensate for the change in flow resistance. The residence time distribution value is used as a constraint condition to limit the temperature adjustment range to prevent excessive thermal degradation of wood flour.

[0013] The improved alternating gray wolf algorithm solution process includes an initialization parameter population stage and an alternating iterative solution stage. The initialization parameter population stage uses the screw speed, feeding speed, mixing section temperature, and zinc stearate external lubricant addition amount as decision variables.

[0014] The first subgame model establishes an upper-level model for thermal stability control, aiming to minimize the degree of wood flour carbonization, and a lower-level model, aiming to maximize plasticization uniformity. These two models achieve synergistic optimization through temperature field distribution uniformity as a coupling term. The second subgame model establishes a lower-level model for improving flowability, aiming to minimize melt viscosity increase, and a higher-level model, aiming to maximize extrusion efficiency. These two models achieve synergistic optimization through energy consumption rate as a coupling term. The first and second upper-level models link the objectives of thermal degradation control and flowability improvement through wood flour thermal stability margin, while the first and second lower-level models link the objectives of plasticization quality and processing efficiency through screw conveying capacity as a coupling term.

[0015] In the alternating iterative solution stage, a two-layer alternating optimization strategy is used to process the two sub-game models respectively. The first round of iteration optimizes the first sub-game model, while fixing the decision variables related to the second sub-game model, namely the amount of zinc stearate external lubricant added and the feeding speed.

[0016] The improved alternating gray wolf algorithm designs the convergence factor as an adaptive nonlinear adjustment mode, dynamically adjusting the search step size according to the current gray wolf population diversity. The gray wolf population diversity is obtained by calculating the standard deviation of the positions of all gray wolf individuals.

[0017] Specifically, adjusting the operating conditions of the twin-screw extruder according to the optimized parameters involves reducing the temperature of the mixing section by 5°C to 10°C and shortening the residence time in the high-temperature zone when the degree of carbonization of wood flour exceeds the thermal stability threshold, and increasing the amount of zinc stearate external lubricant by 0.5% to 1% and increasing the screw speed by 10 r / min to 20 r / min when the increase in melt viscosity exceeds the fluidity threshold.

[0018] This invention proposes a parameter optimization method based on a dual-game model. It constructs a first sub-game model for thermal stability control and a second sub-game model for improving flowability. A temperature-controlled flow collaborative balance algorithm is used to monitor the dynamic response relationship between material temperature distribution and extrusion pressure in real time. When a continuous increase in material temperature is detected, the temperature gradient is automatically reduced to prevent wood flour carbonization. When the extrusion pressure decreases, the screw speed is simultaneously increased to compensate for changes in flow resistance. The two sub-game models achieve global coordination through coupling terms such as wood flour thermal stability margin and screw conveying capacity. This ensures that the carbonization control objective and viscosity control objective are mutually constrained rather than independently optimized during the optimization process. An improved alternating gray wolf algorithm is used to iteratively solve for the optimal screw speed matching value and the optimal feeding speed matching value, ensuring maximum improvement in melt flow performance under high filler content without wood flour carbonization. In summary, this invention solves the technical problem mentioned in the background art where the mutual constraint between thermal stability and flowability in the processing of wood flour plastic composites with high filler content makes it difficult to coordinately optimize processing parameters. Attached Figure Description

[0019] Figure 1 This is a flowchart of the method of the present invention.

[0020] Figure 2 The graph shows the relationship between the material temperature distribution and the extrusion pressure over time in the example.

[0021] Figure 3 This is a diagram illustrating the iterative optimization process of the fitness values ​​of the objective function at each layer of the dual-game model in the embodiment.

[0022] Figure 4 This is a comparison chart before and after optimization in the embodiment, including two sub-charts. Figure 4 A shows the processing parameters before and after optimization. Figure 4 B is a graph showing the changes in the evaluation index of the degree of carbonization of wood flour.

[0023] Figure 5 The graph shows the melt pressure and temperature distribution curves inside the wide-channel die in the embodiment. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0025] like Figure 1 The diagram shown is a flowchart of a method for twin-screw compounding and granulation of bio-based composite materials provided by the present invention. This method includes the following steps:

[0026] S01. In the raw material pretreatment stage, the wood flour is sieved to a particle size range of 100 to 200 mesh, and then placed in a drying equipment and dried at a temperature of 80°C to 90°C for 4 to 6 hours to reduce the moisture content to below 2%. The polyethylene resin is screened according to the melt flow rate standard of 5 to 15 g / 10 min and mixed with zinc stearate external lubricant at a mass ratio of 100:2 to 100:4 for later use.

[0027] S02, Gradient temperature control mixing stage: The treated wood flour and polyethylene resin are fed into a twin-screw extruder at a mass ratio of 65:35 to 75:25. The feeding section temperature is set to 100℃ to 120℃, the mixing section temperature to 140℃ to 160℃, the die head section temperature to 150℃ to 170℃, the screw speed is controlled in the range of 200r / min to 300r / min, and the screw configuration with a length-to-diameter ratio of 40 to 48 is selected.

[0028] S03. Real-time monitoring of wood flour thermal degradation parameters and melt flow resistance parameters and data collection. Wood flour thermal degradation parameters include material temperature distribution value and residence time distribution value. Melt flow resistance parameters include extrusion pressure value and torque load value. When the material temperature distribution value shows an upward trend, the extrusion pressure value shows a downward trend, and the residence time distribution value remains relatively stable, the temperature control flow collaborative balance algorithm is activated.

[0029] S04. Based on the dual game model, the processing parameters are optimized. The first sub-game model is established for thermal stability control, with the goal of minimizing the degree of wood flour carbonization and the goal of maximizing plasticization uniformity. The second sub-game model is established for fluidity improvement, with the goal of minimizing the increase in melt viscosity and the goal of maximizing extrusion efficiency. The improved alternating gray wolf algorithm is used to iteratively solve the optimal screw speed matching value and the optimal feeding speed matching value.

[0030] S05. Adjust the operating conditions of the twin-screw extruder according to the optimized parameters. When the degree of carbonization of wood flour exceeds the thermal stability threshold, reduce the temperature of the mixing section by 5°C to 10°C and shorten the residence time in the high-temperature zone. When the increase in melt viscosity exceeds the fluidity threshold, increase the amount of zinc stearate external lubricant by 0.5% to 1% and increase the screw speed by 10 r / min to 20 r / min.

[0031] S06. After the extrudate is formed by the wide runner die, it enters the water cooling system to be cooled to below 40°C. Then, it is cut into cylindrical particles with a length of 2mm to 4mm by a pelletizer. The particles are screened by a vibrating screen, then subjected to quality inspection and packaged for storage.

[0032] The temperature-controlled flow synergistic balance algorithm achieves processing stability control by constructing a dynamic response relationship between material temperature distribution and extrusion pressure. When the material temperature distribution is detected to be continuously rising, the algorithm automatically reduces the temperature gradient between the feeding section and the mixing section. When the extrusion pressure decreases, the algorithm simultaneously increases the screw speed to compensate for changes in flow resistance. The residence time distribution serves as a constraint to limit the temperature adjustment range and prevent excessive thermal degradation of the wood flour. The synergistic adjustment of these three parameters ensures that the composite material avoids carbonization and discoloration of heat-sensitive wood flour while maintaining plasticization quality. The material temperature distribution is obtained in real time by installing temperature sensors in the feeding section, mixing section, and die head section of the twin-screw extruder, with a sampling frequency of 1Hz to 5Hz. The residence time distribution is calculated by adding a tracer to the material and detecting the tracer concentration change curve at the extruder outlet. The tracer is an inorganic pigment with good thermal stability. The extrusion pressure is obtained in real time by installing a pressure sensor at the front end of the die head, with a measurement range of 0MPa to 50MPa. The torque load value is obtained by the twin-screw extruder motor power monitoring system, with a sampling frequency of 1Hz to 5Hz.

[0033] The thermal stability threshold is determined by measuring the color change rate of wood flour under different temperature and time conditions. When the color change rate exceeds 15%, it is considered that the degree of carbonization of the wood flour exceeds the standard, triggering a temperature reduction mechanism. The thermal stability threshold setting needs to be calibrated in conjunction with the type and particle size distribution of wood flour to adapt to different raw material characteristics. The flowability threshold is determined by measuring the apparent viscosity of the melt at a standard shear rate. When the apparent viscosity increases by more than 300% compared to the pure polyethylene matrix, it is considered that the flowability has seriously deteriorated, triggering a lubricant increment mechanism and a speed increase mechanism to synergistically improve processing performance. The color change rate is calculated by measuring the Lab color space difference between the composite material sample and a standard white board using a colorimeter. The color difference value reflects the degree of thermal degradation of the wood flour during processing. The apparent viscosity is measured by a capillary rheometer at a shear rate of 100... Up to 1000 Obtained by measurement within the specified range.

[0034] In the first subgame model, the objective function of the first upper-level model quantifies the correlation between the degree of wood flour carbonization and processing temperature and time. Inputs include the mixing zone temperature, high-temperature zone residence time, and wood flour particle size distribution. The output is a carbonization degree evaluation index. The objective function of the first lower-level model quantifies the correlation between plasticizing uniformity and screw shearing action. Inputs include screw speed, length-to-diameter ratio, and material filling rate. The output is a plasticizing uniformity evaluation index. The two models achieve synergistic optimization through temperature field distribution uniformity as a coupling term. In the second subgame model, the objective function of the second upper-level model quantifies the correlation between melt viscosity increase and wood flour filling amount. Inputs include wood flour mass fraction, zinc stearate external lubricant addition, and melt temperature. The output is a viscosity increase evaluation index. The objective function of the second lower-level model quantifies the correlation between extrusion efficiency and processing rate. Inputs include feeding rate, screw speed, and die pressure. The output is an extrusion efficiency evaluation index. The two models achieve synergistic optimization through energy consumption rate as a coupling term. The first upper-level model and the second upper-level model use the wood flour thermal stability margin as a coupling term to link the two objectives of thermal degradation control and flowability improvement. The first lower-level model and the second lower-level model use the screw conveying capacity as a coupling term to link the two objectives of plasticizing quality and processing efficiency.

[0035] The mixing section temperature value is derived from the mixing section temperature set in step S02, and its range is 140℃ to 160℃. The residence time value in the high-temperature zone is obtained by accumulating the time periods in the residence time distribution where the material temperature exceeds 140℃. The wood flour particle size distribution value is obtained by measuring the particle size distribution of the wood flour after sieving in step S01 using a laser particle size analyzer, and the particle size distribution conforms to the characteristics of a normal distribution. The aspect ratio value is derived from the screw configuration selected in step S02, and its range is 40 to 48. The material filling rate value is obtained by calculating the ratio of the screw channel volume of the twin-screw extruder to the actual material volume, and its range is 30% to 70%. The wood flour mass fraction value is derived from the mass ratio of wood flour to polyethylene resin in step S02. The amount of zinc stearate external lubricant added is derived from the mass ratio of zinc stearate external lubricant to polyethylene resin in step S01, with an initial value range of 2% to 4%, which is dynamically adjusted in step S05 based on the result of the flowability threshold judgment. The melt temperature value is obtained in real time by a melt temperature sensor installed in the die head section. The feeding rate value is set and acquired by the frequency converter of the twin-screw extruder feeding system, with a range of 10 kg / h to 50 kg / h. The die head pressure value is obtained in real time by a pressure sensor installed at the die head inlet.

[0036] The carbonization degree evaluation index is used in step S05 to determine whether the carbonization degree of wood flour exceeds the thermal stability threshold. A higher carbonization degree evaluation index value indicates a worse thermal stability control effect. The plasticization uniformity evaluation index is used to evaluate the mixing effect. A higher plasticization uniformity evaluation index value indicates that the wood flour is more uniformly dispersed in the polyethylene matrix. The viscosity increase evaluation index is used in step S05 to determine whether the melt viscosity increase exceeds the flowability threshold. A lower viscosity increase evaluation index value indicates better flowability retention. The extrusion efficiency evaluation index is used to evaluate processing economy. A higher extrusion efficiency evaluation index value indicates a better ratio of output to energy consumption per unit time. The optimal screw speed matching value and optimal feeding speed matching value are used in step S05 to adjust the operating conditions of the twin-screw extruder.

[0037] The constraints of the first upper-level model include the following: temperature range constraints in the mixing section, upper limit constraints on the residence time in the high-temperature zone, and upper limit constraints on the wood powder carbonization degree evaluation index. The mixing section temperature must be ∈ [140, 160]℃, the residence time in the high-temperature zone must be ≤180s, and the wood powder carbonization degree evaluation index must be ≤15, corresponding to a color change rate of 15%. The constraints of the first lower-level model include the following: screw speed range constraints, material filling rate range constraints, and lower limit constraints on the plasticization uniformity evaluation index. The screw speed must be ∈ [200, 300]r / min, the material filling rate must be ∈ [30, 70]%, and the plasticization uniformity evaluation index must be ≥0.8 normalization standard. The constraints of the second upper-level model include constraints on the range of wood flour mass fraction, the upper limit of the amount of zinc stearate external lubricant added, and the upper limit of the melt viscosity increase evaluation index. The wood flour mass fraction must be ∈ [65, 75], the amount of zinc stearate external lubricant added must be ≤5% to prevent precipitation, and the melt viscosity increase evaluation index must be ≤300, corresponding to a 300% increase in apparent viscosity. The constraints of the second lower-level model include constraints on the range of feeding speed, the upper limit of the die pressure, and the lower limit of the extrusion efficiency evaluation index. The feeding speed must be ∈ [10, 50] kg / h, the die pressure must be ≤40 MPa to prevent die deformation, and the extrusion efficiency evaluation index must be ≥0.6 normalization standard. The temperature field distribution uniformity, as an internal coupling term of the first sub-game model, must satisfy that the standard deviation of the temperature in the feeding section, mixing section, and die head section is ≤10℃. The energy consumption rate, as an internal coupling term of the second sub-game model, must satisfy that the energy consumption per unit output is ≤2kWh / kg. The thermal stability margin of the wood flour, as a coupling term between the first and second upper-level models, must meet a safety margin of ≥20℃. The screw conveying capacity, as a coupling term between the first and second lower-level models, must meet a ratio of actual conveying capacity to theoretical conveying capacity of ≥0.85.

[0038] The reason for adopting the dual-game model is that there is a mutually restrictive relationship between thermal stability and flowability during the processing of wood flour-plastic composites. Increasing the temperature to improve flowability will exacerbate the thermal degradation of wood flour, while decreasing the temperature to protect the thermal stability of wood flour will worsen melt flowability. Traditional single-objective optimization methods are difficult to balance these two contradictions. The dual-game model addresses the contradiction between carbonization control and plasticization uniformity within thermal stability through the first sub-game, and the contradiction between viscosity control and efficiency improvement within flowability through the second sub-game. The two sub-games achieve global coordination through coupling terms, enabling the processing parameters to maximize the improvement of flowability under high filler content while ensuring that the wood flour does not carbonize. The technical effects brought by the dual-game model to the entire solution include: a significant reduction in extrusion pressure and torque load values ​​when the wood flour addition reaches 65% to 75%; a significant improvement in the uniformity of the composite material's appearance and color; stable mechanical properties of the product; a significant reduction in processing energy consumption compared to conventional processes; a significant reduction in mold wear; and effective assurance of production continuity and stability.

[0039] The improved alternating gray wolf algorithm solution process includes an initial parameter population stage and an alternating iterative solution stage. In the initial parameter population stage, screw speed, feeding rate, mixing section temperature, and zinc stearate external lubricant addition are used as decision variables. Within their respective process ranges, 30 to 50 initial solutions are randomly generated as gray wolf individuals. Each gray wolf individual corresponds to a complete set of processing parameters. The fitness value of each gray wolf individual in terms of thermal stability control is calculated using the objective functions of the first upper-level model and the first lower-level model of the first subgame model. Similarly, the fitness value of each gray wolf individual in terms of fluidity improvement is calculated using the objective functions of the second upper-level model and the second lower-level model of the second subgame model.

[0040] In the alternating iterative solution stage, a two-layer alternating optimization strategy is adopted to handle the two sub-game models separately. The first iteration optimizes the first sub-game model, while fixing the decision variables related to the second sub-game model, namely the amount of zinc stearate external lubricant added and the feeding speed. The gray wolf algorithm is used to update the decision variables related to the first sub-game model, namely the mixing section temperature and screw speed. Population evolution is achieved by simulating the hierarchical structure of the gray wolf group, namely the alpha wolf, the second wolf, and the last wolf. The gray wolf with the best fitness value in the first upper-level model is marked as the alpha wolf A, the gray wolf with the best fitness value in the first lower-level model is marked as the second wolf A, and the gray wolf with the third best overall fitness value is marked as the last wolf A. The remaining gray wolves update their own positions according to their positional relationship with the alpha wolf A, the second wolf A, and the last wolf A. A convergence factor is introduced into the position update formula to control the search step size. The convergence factor decreases linearly from 2 to 0 with the number of iterations, realizing the transformation from global exploration to local development. After the first round of iterations, the current optimal solution of the first subgame model is obtained, including the optimal mixing section temperature value and the optimal screw speed value.

[0041] The second iteration optimizes the second subgame model. The decision variables related to the first subgame model—the mixing section temperature and screw speed—are fixed as the optimal solution from the first iteration. The Grey Wolf algorithm is used to update the decision variables related to the second subgame model—the amount of zinc stearate external lubricant added and the feeding rate. The Grey Wolf individual with the best fitness value in the upper-level model is labeled as the Head Wolf B, the Grey Wolf individual with the best fitness value in the lower-level model is labeled as the Second Wolf B, and the Grey Wolf individual with the third best overall fitness value is labeled as the Last Wolf B. The remaining Grey Wolves update their positions based on their positional relationships with Head Wolf B, Second Wolf B, and Last Wolf B. During the update process, a convergence factor is used to control the search step size. After the second iteration, the current optimal solution for the second subgame model is obtained, including the optimal amount of zinc stearate external lubricant added and the optimal feeding rate.

[0042] Two iterations constitute a complete alternating optimization cycle. This cycle is repeated until the convergence criterion is met or the maximum number of iterations (100 to 200) is reached. The convergence criterion is that the improvement in the objective function value of the first subgame model is less than 0.1% and the improvement in the objective function value of the second subgame model is less than 0.1% in five consecutive alternating optimization cycles. When the convergence criterion is met, the final optimal screw speed matching value, optimal feeding speed matching value, and the corresponding optimal mixing section temperature value and optimal zinc stearate external lubricant addition amount are output. The alternating optimization strategy ensures that the first and second subgame models transmit information through the fixing and updating of decision variables during the optimization process, reflecting the mutual influence and coordination between the two subgames in the dual game.

[0043] The improvement lies in the fact that the traditional gray wolf algorithm, which uses a linear decreasing strategy for its convergence factor, is prone to getting trapped in local optima. The improved alternating gray wolf algorithm designs the convergence factor with an adaptive nonlinear adjustment mode. It dynamically adjusts the search step size based on the current gray wolf population diversity. Gray wolf population diversity is obtained by calculating the standard deviation of the positions of all gray wolves. When the standard deviation of gray wolf population diversity is greater than the threshold of 0.3, it is considered that the diversity is high, and the convergence factor is increased to 1.5 times to maintain the exploration range. When the standard deviation of gray wolf population diversity is less than the threshold of 0.1, it is considered that the diversity is low, and the convergence factor is decreased to 0.5 times to accelerate the convergence process. This adaptive adjustment mechanism significantly improves the solution quality and convergence speed of the improved alternating gray wolf algorithm in multi-objective coupled optimization problems. The adaptive nonlinear adjustment mode evaluates the gray wolf population diversity and updates the convergence factor value every 10 iterations.

[0044] The degree of wood powder carbonization was calculated by measuring the color difference value of the composite material sample. The color difference value was obtained by measuring the Euclidean distance between the sample and the standard white board in the Lab color space using a colorimeter. A higher degree of wood powder carbonization indicates a more severe degree of thermal degradation of the wood powder during processing. The degree of wood powder carbonization was used in step S05 to compare with the thermal stability threshold to determine whether the mixing section temperature needs to be reduced. Plasticization uniformity was calculated by analyzing the uniformity of the component distribution across the cross-section of the extruded material. The dispersion state of wood powder in the polyethylene matrix was observed using a scanning electron microscope, and the standard deviation of the wood powder particle spacing was statistically analyzed. A smaller standard deviation indicates higher plasticization uniformity. The melt viscosity increase was calculated by measuring the rheological properties of the composite material melt. The apparent viscosity at different shear rates was measured using a capillary rheometer and compared with that of the pure polyethylene matrix. A smaller melt viscosity increase indicates better fluidity retention. The melt viscosity increase was used in step S05 to compare with the fluidity threshold to determine whether the amount of zinc stearate external lubricant needs to be increased. Extrusion efficiency is obtained by calculating the ratio of output to energy consumption per unit time. Output is calculated by dividing the mass of the pellets after weighing and cutting by the time, and energy consumption is obtained by the motor power of the integral twin-screw extruder.

[0045] In this model, temperature field uniformity, as an internal coupling term of the first sub-game model, is obtained by measuring the material temperature at different axial positions of the twin-screw extruder. Five to eight temperature measurement points are arranged in the feeding section, mixing section, and die head section, respectively. The temperature field uniformity value is characterized by the standard deviation of the temperatures at all measurement points. A more uniform temperature distribution indicates more consistent heating of the wood flour, a lower risk of carbonization, and more stable plasticization quality. Temperature field uniformity ensures that the carbonization control objective of the first upper-level model and the plasticization uniformity objective of the first lower-level model are coordinated rather than independently optimized during the optimization process. Energy consumption rate, as an internal coupling term of the second sub-game model, is obtained by measuring the ratio of motor power to output of the twin-screw extruder. Motor power is collected in real time using a power analyzer. A lower energy consumption rate indicates lower melt flow resistance and higher processing efficiency. Energy consumption rate ensures that the viscosity control objective of the second upper-level model and the efficiency improvement objective of the second lower-level model are coordinated during the optimization process. The wood flour thermal stability margin, as a coupling term between the first and second upper-level models, is defined as the difference between the critical temperature at which wood flour begins to carbonize and the actual temperature in the mixing section. The critical temperature at which wood flour begins to carbonize is obtained by measuring the temperature corresponding to the point of abrupt change in the mass loss rate of the wood flour sample during the heating process using a thermogravimetric analyzer. A larger wood flour thermal stability margin indicates that the processing process is further away from the risk of carbonization. The wood flour thermal stability margin enables the carbonization control of the first sub-game model and the viscosity control of the second sub-game model to be synergistically optimized at the global level, avoiding carbonization caused by simply reducing viscosity and resulting in excessively high temperatures. The screw conveying capacity, as a coupling term between the first and second lower-level models, is obtained by measuring the material conveying rate of the screw at different speeds. The material conveying rate is calculated by continuously weighing the extrusion volume per unit time at the extruder outlet. A stronger screw conveying capacity indicates a better match between the plasticizing and extrusion processes. The screw conveying capacity enables the plasticizing uniformity of the first sub-game model and the efficiency improvement of the second sub-game model to be synergistically optimized at the global level.

[0046] The wide-channel die features a cross-sectional area 30% to 50% larger than that of the standard die. This increased channel width reduces melt flow resistance within the die, lowering extrusion pressure and shear heat generation. It also prevents material stagnation due to flow obstruction, which could lead to localized overheating and wood powder carbonization. The wide-channel die design, combined with a gradient temperature control strategy and the addition of zinc stearate as an external lubricant, addresses the issue of deteriorated flowability under high filler loads. The cross-sectional area is calculated by measuring the width and height of the internal channels of the wide-channel die. The standard die has a cross-sectional area of ​​200 mm. Up to 300 The cross-sectional area of ​​the wide-channel mold head is 260. Up to 450 .

[0047] The specific implementation methods of the above steps are described in detail below.

[0048] The specific implementation of step S01 involves using a mechanical vibrating sieve to classify the raw wood flour by particle size. The sieve mesh size is set within the range of 100 to 200 mesh, controlling the wood flour particle size to be between 75 μm and 150 μm. This particle size range ensures the uniform dispersion of the wood flour in the polyethylene matrix while avoiding stress concentration caused by excessively large particles. The sieved wood flour is then transferred to a hot air circulating oven for dehydration. The internal temperature of the oven is stably controlled at 80°C to 90°C, and the treatment time lasts for 4 to 6 hours. The moisture content of the wood flour is monitored in real time using an infrared moisture analyzer until it drops below 2%. The purpose of the drying treatment is to eliminate free water and some bound water in the wood flour, preventing moisture vaporization during subsequent high-temperature processing, which could lead to porosity defects in the product, and also avoiding water... The deterioration of the compatibility between wood flour and polyethylene interface caused by the reaction of the resin was addressed by screening the polyethylene resin according to the melt flow rate standard. The flowability index of the resin was measured by a melt flow rate meter under the conditions of 190℃ and 2.16kg load. The resin grades with a melt flow rate of 5g / 10min to 15g / 10min were selected. This flowability range can meet the processing flow requirements of the high-filled system and ensure that the product has sufficient mechanical strength. Zinc stearate external lubricant was added at a ratio of 2% to 4% of the polyethylene resin mass. Zinc stearate and polyethylene resin were fully premixed for 3 to 5 minutes by a high-speed mixer. The mechanism of action of the external lubricant is to form a low friction coefficient lubricating layer between the melt and the metal wall, reduce the adhesion tendency of the material on the screw and barrel inner wall, and reduce flow resistance and shear heat generation.

[0049] The specific implementation of step S02 involves accurately metering the dried wood flour and premixed polyethylene resin at a mass ratio of 65:35 to 75:25 using a weighing feeding system and then feeding them into the hopper of a twin-screw extruder. The twin-screw extruder adopts a co-rotating parallel screw structure with a length-to-diameter ratio of 40 to 48 to provide sufficient mixing length and residence time. The temperature of the feeding section is set to 100°C to 120°C, which allows the polyethylene resin to begin to soften but not completely melt, maintaining the solid conveying characteristics of the material and ensuring a stable and continuous feeding process. The temperature of the mixing section is set to 140°C to 160°C, which allows the polyethylene to fully melt and coat and disperse the wood flour, while remaining below the thermal degradation temperature threshold of the wood flour to avoid carbonization. The die head section temperature is set at 150℃ to 170℃, slightly higher than the mixing section temperature, to compensate for heat loss at the die head and reduce melt viscosity for easier extrusion molding. The screw speed is controlled within the range of 200r / min to 300r / min by a variable frequency motor drive system. The selection of the speed needs to balance the shear mixing effect and residence time. Too low a speed will result in uneven dispersion, while too high a speed will shorten the residence time and generate excessive shear heat. The gradient temperature control design principle is to establish a temperature increase or stable distribution based on the processing requirements of the material in different areas of the screw, avoiding thermal stress and processing instability caused by sudden temperature changes. The three-stage temperature control is achieved through independent heating and cooling systems. Each temperature zone is equipped with an electric heating coil and an air-cooled or water-cooled device, and the temperature deviation is controlled within ±2℃.

[0050] The specific implementation of step S03 involves installing K-type thermocouple temperature sensors in the feeding section, mixing section, and die head section of a twin-screw extruder. The sensor probes are inserted into the material flow area between the screw and barrel. The data acquisition frequency is set to 1Hz to 5Hz. The material temperature distribution value is obtained through statistical analysis of multi-point temperature data. A moving average filtering algorithm is used to eliminate interference from instantaneous temperature fluctuations. The residence time distribution value is determined by adding 0.1% to 0.3% by mass of an inorganic pigment tracer, such as titanium dioxide or iron oxide, to the material at the feeding port. The tracer has good thermal stability and does not decompose at the processing temperature. At the extruder outlet, an online spectrophotometer continuously monitors the color concentration change of the extruded material. Based on the tracer concentration-time response curve, the average residence time and its distribution width within the extruder are calculated. The residence time distribution value reflects the length of time the material is exposed to the high-temperature zone and is a key parameter for assessing the risk of thermal degradation. The extrusion pressure value is obtained through a piezoresistive pressure sensor installed at the die head flange. The device collects data in real time, with a measurement range covering 0MPa to 50MPa. The pressure signal is transmitted to the data acquisition system through an amplifier and an analog-to-digital converter. Changes in extrusion pressure reflect melt flow resistance and material rheological properties. A decrease in pressure usually means a decrease in melt viscosity or insufficient material filling. The torque load value is obtained through the power monitoring module of the main motor of the twin-screw extruder. The motor torque is directly related to power and speed. The torque value characterizes the driving force required for the screw to overcome the viscous resistance of the material. When the material temperature distribution value shows a continuous upward trend, the extrusion pressure value shows a downward trend, and the residence time distribution value remains relatively stable within the set range, the system determines that the processing process has entered a state that requires intervention. At this time, the temperature control flow synergistic balance algorithm is activated to adjust the parameters. This judgment logic is based on the coupling relationship between wood flour thermal degradation and melt rheological behavior. Increased temperature will reduce melt viscosity and cause a decrease in pressure, but at the same time, it will increase the risk of wood flour carbonization. The goal of the synergistic balance algorithm is to control the temperature to avoid thermal degradation while ensuring fluidity.

[0051] The specific implementation of step S04 involves constructing a dual-game model framework to describe the multi-objective conflict and coordination relationship in the processing parameter optimization problem. The first sub-game model focuses on thermal stability control. The first upper-level model aims to minimize the degree of wood flour carbonization. Input parameters include the mixing section temperature (range 140℃ to 160℃), the high-temperature zone residence time (obtained by accumulating the time period when the material temperature exceeds 140℃), and the wood flour particle size distribution measured by a laser particle size analyzer. The output is a carbonization degree evaluation index, calculated by measuring the Lab color space difference between the composite material sample and a standard white board using a colorimeter. A higher value indicates more severe carbonization. The first lower-level model aims to maximize plasticization uniformity. Input parameters include the screw speed (range 200 r / min to 300 r / min), length... The aspect ratio is 40 to 48, and the material filling rate is calculated as 30% to 70% by the ratio of screw channel volume to actual material volume. The output is a plasticization uniformity evaluation index, which is obtained by observing the dispersion state of wood flour in the polyethylene matrix using scanning electron microscopy and statistically analyzing the standard deviation of particle spacing. The two models achieve synergy through temperature field distribution uniformity as a coupling term. Temperature field uniformity is characterized by the standard deviation of temperature in each temperature zone and must be less than 10℃. The second sub-game model focuses on improving fluidity, with the second upper-level model aiming to minimize the increase in melt viscosity. Input parameters include a wood flour mass fraction of 65% to 75%, an initial zinc stearate external lubricant addition of 2% to 4%, and melt temperature values ​​collected by a temperature sensor in the die head section. The output is a viscosity increase evaluation index, which is obtained by a capillary rheometer at a shear rate of 100. Up to 1000 The apparent viscosity was measured within a certain range and compared with that of pure polyethylene. The second lower-level model aimed to maximize extrusion efficiency. Input parameters included feed rate (10 kg / h to 50 kg / h), screw speed, and die pressure collected by a die inlet pressure sensor. The output was an extrusion efficiency evaluation index, calculated as the ratio of output per unit time to energy consumption. The two models were coupled using energy consumption rate as a coupling term, requiring energy consumption per unit output to not exceed 2%. The two subgames are globally correlated through the wood flour thermal stability margin and screw conveying capacity as cross-level coupling terms. The thermal stability margin is defined as the difference between the critical carbonization temperature of wood flour and the actual mixing temperature must be greater than 20℃. The screw conveying capacity requires the ratio of actual conveying capacity to theoretical conveying capacity to be no less than 0.85. An improved alternating gray wolf algorithm is used to iteratively solve the dual-game model. The algorithm first randomly generates 30 to 50 sets of initial parameter combinations within the feasible region of each decision variable as a gray wolf population. Each individual corresponds to a complete set of processing parameters, including screw speed, feeding speed, mixing section temperature, and zinc stearate external lubricant addition. The fitness value of each individual under the four objective functions is calculated. The alternating iteration process is carried out in two rounds. In the first round, the relevant variables of the second subgame, namely the lubricant addition and feeding speed, are fixed, and only the relevant variables of the first subgame, namely the mixing temperature and screw speed, are optimized. By simulating the gray wolf hierarchy, the optimal individual of the first upper-level model is marked as the alpha wolf. The optimal individual in the layer model is labeled as the second-best wolf, and the individual with the third-best overall fitness is labeled as the last wolf. The remaining individuals update their positions based on their positional relationship with the three leader wolves. In the second round, the variables related to the first subgame are fixed as the optimal solution of the previous round, and the variables related to the second subgame are optimized. Similarly, a hierarchical structure of leader wolf B, second-best wolf B, and last wolf B is established for population updates. The two rounds constitute an alternating optimization cycle, which is repeated until the objective function improvement of the two subgames is less than 0.1% in five consecutive cycles or the maximum number of iterations (100 to 200) is reached. The improvement lies in the adaptive nonlinear adjustment mode of the convergence factor. Every 10 iterations, the standard deviation of the population position is calculated to assess diversity. When the standard deviation is greater than 0.3, the convergence factor is increased to 1.5 times to maintain exploration. When the standard deviation is less than 0.1, the convergence factor is reduced to 0.5 times to accelerate convergence. Finally, the optimal screw speed matching value, the optimal feeding speed matching value, the corresponding optimal mixing section temperature value, and the optimal amount of zinc stearate external lubricant added are output.

[0052] The specific implementation of step S05 involves adjusting the operating conditions of the twin-screw extruder based on the parameter combination optimized in step S04. The optimal screw speed matching value is set to the frequency converter drive system via a programmable logic controller (PLC), and the optimal feeding speed matching value is set to the screw feeder speed control module. The optimal mixing section temperature value and the optimal zinc stearate external lubricant addition value are updated to the temperature control system and the batching system. During production, two key indicators—the degree of wood flour carbonization and the increase in melt viscosity—are continuously monitored. The degree of wood flour carbonization is calculated by real-time measurement of the color difference value of the extruded material using an online colorimeter. When the color change rate corresponding to the color difference value exceeds 15%, the degree of carbonization is determined to exceed the thermal stability threshold. At this point, the control system... The system automatically reduces the temperature of the mixing section by 5°C to 10°C. At the same time, it shortens the residence time of the material in the high-temperature zone by increasing the screw speed or reducing the feeding speed. The increase in melt viscosity is estimated by an online rheometer or based on the trend of extrusion pressure and torque. When the viscosity increase exceeds 300%, it is determined that the fluidity has seriously deteriorated and exceeded the fluidity threshold. At this time, the control system automatically increases the amount of zinc stearate external lubricant by 0.5% to 1% and increases the screw speed by 10 r / min to 20 r / min to enhance the shearing action and reduce the apparent viscosity. The dual adjustment mechanism achieves dynamic stability of the processing through a feedback control loop, ensuring that thermal degradation is avoided and good processing fluidity is maintained even when the wood flour filling content is high.

[0053] The specific implementation of step S06 involves extruding the mixed melt through a wide runner die. The cross-sectional area of ​​the wide runner die is designed to be 260. Up to 450 Compared to the usual 200 Up to 300 The width of the flow channel is increased by 30% to 50%, which significantly reduces the melt velocity and shear rate within the die, reducing pressure loss and shear heat generation caused by flow resistance. This prevents overheating due to localized retention, which could lead to wood powder carbonization. The extruded melt strips enter a water cooling tank for rapid cooling, with the cooling water temperature controlled between 15°C and 25°C. The residence time of the material in the water cooling system is determined based on the strip diameter and extrusion speed, ensuring that the core temperature of the material drops below 40°C to achieve sufficient rigidity for subsequent pelletizing. The cooled strips are then conveyed at a stable linear speed to a rotary blade pelletizer via a traction device. The pelletizer blade speed is synchronized with the traction speed, and the length of the cut pellets is controlled within the range of 2mm to 4mm. The pellets are regular cylindrical in shape. After cutting, the pellets are sent to a vibrating screen through an air conveying system for screening to remove unqualified pellets that are too long, too short, or sticking together. Qualified pellets are sampled for quality testing, including indicators such as appearance, color uniformity, density, and mechanical properties. After passing the test, the pellets are centrifuged, dehydrated, and dried with hot air to remove residual moisture from the surface. Finally, they are packaged by an automatic packaging system according to the specified weight and heat-sealed in bags. The finished pellets are stored in a dry and ventilated warehouse to avoid moisture and ultraviolet radiation.

[0054] It should be noted that the key technical ideas of this invention include the application of a temperature-controlled flow synergistic balance algorithm, an optimization framework of a dual game model, and a solution strategy of an improved alternating gray wolf algorithm. The temperature-controlled flow synergistic balance algorithm establishes a synergistic mechanism of temperature control and flowability compensation by real-time monitoring of the dynamic changes in three parameters: material temperature distribution, extrusion pressure, and residence time. Compared with traditional single-parameter control methods, this algorithm can find the optimal balance between the heat sensitivity of wood flour and melt flowability. This prevents the problem of wood flour carbonization and discoloration caused by excessively high temperatures, and avoids the sharp increase in flow resistance and extrusion difficulties caused by excessively low temperatures, ensuring the processing stability and product quality consistency of high-filler-content wood-plastic composites during continuous production. The dual-game model decomposes thermal stability control and flowability improvement into two relatively independent yet interconnected sub-problems. Each sub-problem contains a game relationship between the upper and lower objectives. A multi-level optimization constraint network is established through four coupling terms: temperature field distribution uniformity, energy consumption rate, wood flour thermal stability margin, and screw conveying capacity. Compared to traditional weighted multi-objective optimization methods, the dual-game model can more accurately describe the nonlinear coupling relationship and objective conflict characteristics between processing parameters, making the optimization results more consistent with the complex constraints in actual production and significantly improving the comprehensive performance and processing economy of composite materials. The improved alternating gray wolf algorithm introduces an adaptive nonlinear convergence factor adjustment mechanism, dynamically adjusting the search step size according to population diversity. This effectively overcomes the defect of the traditional gray wolf algorithm being prone to getting trapped in local optima. In multi-objective coupled optimization problems of the dual-game model, it exhibits faster convergence speed and higher solution accuracy. Compared to traditional intelligent optimization methods such as genetic algorithms and particle swarm optimization, the improved algorithm has stronger global search capabilities and local refinement capabilities when dealing with multi-level nested optimization problems.

[0055] Specifically, the principle of this invention is as follows: The fundamental reason why this invention can solve the technical problem is that it treats thermal stability control and fluidity improvement as an interrelated game process rather than isolated optimization objectives. The first sub-game model establishes an internal balance relationship for thermal stability through a first upper-level model that minimizes the degree of wood flour carbonization and a first lower-level model that maximizes plasticization uniformity. The second sub-game model establishes an internal balance relationship for fluidity through a second upper-level model that minimizes the increase in melt viscosity and a second lower-level model that maximizes extrusion efficiency. The two sub-games are coupled by a wood flour thermal stability margin term to ensure that the optimization direction of reducing viscosity does not lead to excessive temperature. The carbonization critical value is used to ensure the matching of the plasticizing and extrusion processes through the screw conveying capacity coupling term. The temperature control and flow synergistic balance algorithm automatically adjusts the temperature gradient and screw speed according to the dynamic response relationship of the real-time collected material temperature distribution, extrusion pressure, and residence time distribution, so that the processing parameters are always in the synergistic balance region of thermal stability and flowability. The improved alternating gray wolf algorithm avoids getting trapped in local optima by adaptively adjusting the convergence factor nonlinearly. The alternating iterative optimization strategy ensures that the first and second subgames achieve information transmission during the fixing and updating of decision variables, thereby enabling the multi-objective coupled optimization problem to converge to the global optimum.

[0056] The following is a specific embodiment 1 of the present invention. The specific implementation of steps S01-S03, S05 and S06 in this embodiment 1 is the same as that described above, and will not be repeated in detail here. Only the specific calculation in step S04 is described below.

[0057] The specific implementation of step S04 involves constructing a dual-game model framework to describe the multi-objective conflict and coordination relationship in the processing parameter optimization problem. The first sub-game model focuses on thermal stability control, and the objective function of the first upper-level model is expressed as follows:

[0058] ;

[0059] In the formula, This is a dimensionless parameter used to evaluate the degree of carbonization. This refers to the temperature value of the mixing section, in °C, ranging from 140 °C to 160 °C. The lower limit of the temperature for the mixing section is set at 140℃. This is the upper limit of the temperature for the mixing section, and its value is 160℃. This is the residence time value in the high-temperature zone, in seconds, obtained by accumulating the time period during which the material temperature exceeds 140℃; This is the minimum dwell time in the high-temperature zone, with a default value of 60 seconds. The maximum time allowed to stay in the high-temperature zone is limited to 180 seconds. The particle size distribution of wood flour is expressed in μm and was determined by a laser particle size analyzer. The lower limit of wood flour particle size is set at 75 μm. This represents the upper limit of wood flour particle size, with a value of 150 μm. The coupling weighting coefficient for the uniformity of the temperature field distribution is a dimensionless parameter with an empirical value of 0.3 to 0.5. This is a dimensionless parameter used to evaluate the uniformity of the temperature field distribution. The coefficient for coupling weighting of the thermal stability margin of wood flour is a dimensionless parameter with an empirical value of 0.2 to 0.4. This is a dimensionless parameter representing the thermal stability margin of wood flour. This is the thermal degradation fluctuation error term, a dimensionless parameter with an empirical value of 0.02 to 0.05.

[0060] Among them, the evaluation value of temperature field distribution uniformity The statement is as follows:

[0061] ;

[0062] In the formula, The standard deviation of temperature in the feeding section, mixing section, and die head section is given in °C. The calculation formula is as follows: ,in This represents the total number of temperature measuring points, ranging from 15 to 24. For the first The temperature values ​​at each measuring point, in °C, are obtained in real time by temperature sensors placed in each temperature zone. The average temperature at all measuring points is expressed in °C. The calculation formula is as follows: ; This is the upper limit of the temperature standard deviation constraint, with a value of 10℃.

[0063] Wood flour thermal stability margin evaluation value The statement is as follows:

[0064] ;

[0065] In the formula, The critical temperature at which wood flour begins to carbonize is measured in °C. It is obtained by measuring the temperature at which the mass loss rate of a wood flour sample changes abruptly during the heating process using a thermogravimetric analyzer, and is typically taken in the range of 180 °C to 200 °C.

[0066] The objective function of the first lower-level model is expressed as follows:

[0067] ;

[0068] In the formula, It is a dimensionless parameter used to evaluate the uniformity of plasticization. This is the screw speed value, in units of... The value range is 200. Up to 300 ; This is the lower limit of the screw speed, with a value of 200. ; This is the upper limit of the screw speed, set to 300. ; The aspect ratio is a dimensionless parameter, ranging from 40 to 48. The lower limit for the aspect ratio is 40; The upper limit for the aspect ratio is 48; This is the material filling rate value, a dimensionless parameter, ranging from 30% to 70%, obtained by calculating the ratio of the screw channel volume of the twin-screw extruder to the actual material volume; This is the lower limit of the material filling rate, with a value of 30%. This represents the upper limit of the material filling rate, with a value of 70%. The coupling weighting coefficient for the uniformity of the temperature field distribution is a dimensionless parameter with an empirical value of 0.3 to 0.5. The screw conveying capacity coupling weighting coefficient is a dimensionless parameter with an empirical value of 0.2 to 0.4. This is a dimensionless parameter representing the screw conveying capacity evaluation value. This is the plasticizing fluctuation error term, a dimensionless parameter with an empirical value of 0.01 to 0.03.

[0069] Among them, the evaluation value of screw conveying capacity The statement is as follows:

[0070] ;

[0071] In the formula, This represents the actual conveying volume, in units of... It is calculated by continuously weighing the extrusion volume per unit time at the extruder outlet; Theoretical transport capacity, unit: The calculation formula is: ,in The average density of the material, in units of The density of the composite material melt is obtained through experimental measurement, and is typically taken as 900. Up to 1100 , The diameter of the screw is given in units of 1000 mm. The value is determined based on the equipment parameters and is typically set to 40. Up to 60 , The depth of the screw groove, in units of The value is determined based on the screw's geometric parameters, and is typically taken as 6. Up to 12 .

[0072] The objective function of the second upper-level model in the second subgame model is expressed as follows:

[0073] ;

[0074] In the formula, This is a viscosity increase evaluation index, a dimensionless parameter; This is a dimensionless parameter representing the mass fraction of wood flour, ranging from 65% to 75%. The lower limit for the mass fraction of wood flour is set at 65%. This represents the upper limit of the wood flour mass fraction, with a value of 75%. The value represents the amount of zinc stearate external lubricant added, a dimensionless parameter, with an initial range of 2% to 4%; This is the lower limit for the addition of zinc stearate as an external lubricant, initially set at 2%. The upper limit for the amount of zinc stearate external lubricant added is set at 5%. This is the melt temperature value, in °C, which is acquired in real time by the melt temperature sensor in the die head section. This is the lower limit of the melt temperature, and its value is 150℃. This is the upper limit of the melt temperature, and its value is 170℃. The energy consumption rate coupling weighting coefficient is a dimensionless parameter with an empirical value of 0.2 to 0.4. This is an evaluation value for energy consumption rate, a dimensionless parameter; The coefficient for coupling weighting of the thermal stability margin of wood flour is a dimensionless parameter with an empirical value of 0.3 to 0.5. This is the rheological fluctuation error term, a dimensionless parameter with an empirical value of 0.03 to 0.06.

[0075] Among them, the energy consumption rate evaluation value The statement is as follows:

[0076] ;

[0077] In the formula, The power rating of the twin-screw extruder motor is given in units of [unit missing]. Data is collected in real time using a power analyzer; Processing time, in hours, is recorded using a production process timer; For output, the unit is The mass of the pellets is calculated by weighing them after pelletizing. The upper limit of the energy consumption rate constraint is set to 2. .

[0078] The objective function of the second lower-level model is expressed as follows:

[0079] ;

[0080] In the formula, This is a dimensionless parameter used to evaluate extrusion efficiency. This is the feeding speed value, in units of... The value range is 10. Up to 50 ; The lower limit for the feeding rate is set to 10. ; The maximum feeding speed is set to 50. ; This is the screw speed value, in units of... The value range is 200. Up to 300 ; This is the lower limit of the screw speed, with a value of 200. ; This is the upper limit of the screw speed, set to 300. ; This is the die head pressure value, in MPa, which is acquired in real time by a pressure sensor at the die head inlet. This is the lower limit of the die head pressure, typically set to 5 MPa. The upper limit constraint for the die head pressure is set at 40 MPa; The energy consumption rate coupling weighting coefficient is a dimensionless parameter with an empirical value of 0.2 to 0.4. The screw conveying capacity coupling weighting coefficient is a dimensionless parameter with an empirical value of 0.3 to 0.5. This is the efficiency fluctuation error term, a dimensionless parameter with an empirical value of 0.02 to 0.04.

[0081] The improved alternating gray wolf algorithm's position update formula is expressed as follows:

[0082] ;

[0083] In the formula, For the first The first gray wolf individual in the... The position vector after each iteration contains four decision variables: screw speed, feeding speed, mixing section temperature, and amount of zinc stearate external lubricant added. The individual index of the gray wolf, with a value ranging from 1 to... ; This represents the current iteration number; For the first The position vector of the alpha wolf in the next iteration represents the position of the gray wolf individual with the best fitness value in the first upper-level model. For the first The position vector of the second wolf in the next iteration represents the position of the gray wolf individual with the best fitness value in the first lower-level model. For the first The position vector of the last wolf in the next iteration represents the position of the gray wolf with the third highest overall fitness value; The convergence factor is a dimensionless parameter. For the first The distance vector between each individual and the leader wolf is calculated using the following formula: ,in This is a random coefficient vector, with each component taking values ​​from 0 to 2, generated using a uniformly distributed random number generator. For the position vector of the leader wolf, use , , The weighted average was obtained. For the first The first gray wolf individual in the... The current position vector in the next iteration This represents the Euclidean norm.

[0084] The formula for adjusting the adaptive convergence factor is expressed as follows:

[0085] ;

[0086] In the formula, This is the initial convergence factor, which defaults to 1; This represents the maximum number of iterations, ranging from 100 to 200. The standard deviation of gray wolf population diversity is a dimensionless parameter, calculated using the following formula: ,in The number of individuals in the gray wolf population, ranging from 30 to 50. For the first The average vector of the positions of all individual gray wolves in the next iteration is calculated using the following formula: .

[0087] It should be explained that the specific implementation of the temperature-controlled flow collaborative balance algorithm is to construct a dynamic response relationship between the material temperature distribution value and the extrusion pressure value. When it is detected that the material temperature distribution value continues to rise and the extrusion pressure value decreases, the temperature-controlled flow collaborative balance algorithm activates the parameter adjustment mechanism. The temperature gradient adjustment amount is described as follows:

[0088] ;

[0089] In the formula, This represents the adjustment amount of the temperature gradient between the feeding section and the mixing section, in °C. A negative value indicates a reduction in the temperature gradient. Temperature response coefficient, unit: The experience value is 5 to 10; This is the change in the average temperature of the material, expressed in °C, and is calculated as the difference between the average values ​​of temperature sensor data from multiple points at the current time and the previous time. The sampling time interval is in seconds and ranges from 0.2 seconds to 1 second. The rate of change of the average temperature of the material over time, in units of ; This is a reference value for extrusion pressure, in MPa. The value is the average extrusion pressure under normal operating conditions, typically ranging from 15MPa to 25MPa. This is the current extrusion pressure value, in MPa, which is acquired in real time by a pressure sensor at the front end of the die head.

[0090] The screw speed compensation amount is described as follows:

[0091] ;

[0092] In the formula, This is the screw speed compensation amount, in units of... A positive value indicates an increase in screw speed; This is the speed response coefficient, in units of... The experience value is 20 to 40; The residence time constraint factor is a dimensionless parameter that ensures the residence time does not exceed the upper limit constraint to prevent excessive thermal degradation of wood flour.

[0093] The carbonization degree evaluation index equation is obtained by multiplying the three key parameters—temperature in the mixing section, residence time in the high-temperature zone, and particle size distribution of wood flour—after normalizing them, and by introducing a coupling term for the uniformity of temperature field distribution. Coupled with wood flour thermal stability margin term The equation includes a thermal degradation fluctuation error term, enabling a quantitative assessment of the thermal stability of wood flour. The temperature field distribution uniformity coupling term links the carbonization control target with the consistency of the temperature distribution. The more non-uniform the temperature field, the larger the carbonization degree evaluation index, indicating a higher risk of thermal degradation. The wood flour thermal stability margin coupling term links the carbonization control target with the safety margin between the actual temperature and the critical temperature. The smaller the margin, the larger the carbonization degree evaluation index, indicating that the carbonization threshold is approaching. This equation considers the physical law that the higher the temperature, the longer the residence time, the larger the particle size, the more non-uniform the temperature distribution, and the smaller the thermal stability margin, the more severe the carbonization. By minimizing this objective function, the risk of thermal degradation of wood flour during processing can be effectively reduced, resulting in a significant improvement in the uniformity of the composite material's appearance color and the absence of obvious browning or black spot defects in the product.

[0094] The plasticization uniformity evaluation index equation is obtained by multiplying the three parameters of screw speed, length-to-diameter ratio, and material filling rate after normalization, and subtracting the coupling term of temperature field distribution uniformity. Coupled with screw conveyor capacity The plasticizing fluctuation error term reflects the influence of screw shearing on the dispersion of wood powder. Among them, the temperature field distribution uniformity coupling term associates the plasticizing uniformity target with the consistency of temperature distribution. When the temperature field is more uniform, the larger the plasticizing uniformity evaluation index, the better the mixing effect. The screw conveying capacity coupling term associates the plasticizing uniformity target with the matching degree of conveying capacity. When the conveying capacity is stronger, the larger the plasticizing uniformity evaluation index, the better the plasticizing and conveying are coordinated. This equation reflects the processing law that the higher the rotation speed, the larger the length-to-diameter ratio, the more moderate the filling rate, the more uniform the temperature distribution, and the stronger the screw conveying capacity, the more uniform the plasticizing. By maximizing this objective function, the dispersion of wood powder in the polyethylene matrix is ​​more uniform, the mechanical properties of the product remain stable, and the fluctuation range of tensile strength and impact strength is controlled within 5%.

[0095] The viscosity increase evaluation index equation is obtained by multiplying the normalized parameters of wood flour mass fraction, lubricant addition amount, and melt temperature, and introducing an energy consumption rate coupling term. Coupled with wood flour thermal stability margin term The rheological fluctuation error term quantitatively describes the degree of melt flowability deterioration under high filler content. Among them, the energy consumption rate coupling term links the viscosity control target with energy economy. When the energy consumption rate is higher, the viscosity increase evaluation index is larger, indicating that the flowability improvement comes at the cost of excessive energy consumption. The wood flour thermal stability margin coupling term links the viscosity control target with the risk of thermal degradation. When the margin is smaller, the viscosity increase evaluation index is larger, indicating that simply reducing viscosity may cause carbonization. This equation reflects the rheological characteristics that the higher the wood flour content, the less lubricant, the lower the melt temperature, the higher the energy consumption rate, and the smaller the thermal stability margin, the greater the viscosity increase. By minimizing this objective function, the extrusion pressure and torque load values ​​are significantly reduced, and the processing energy consumption is reduced by 15% to 20% compared with conventional processes.

[0096] The extrusion efficiency evaluation index equation is calculated by multiplying the three parameters of feeding rate, screw speed, and die pressure after normalization, and then subtracting the energy consumption rate coupling term. Coupled with screw conveyor capacity In addition to the efficiency fluctuation error term, the balance between processing rate and energy consumption is comprehensively evaluated. Among them, the energy consumption rate coupling term links the extrusion efficiency target with energy economy. When the energy consumption rate is lower, the larger the extrusion efficiency evaluation index, the higher the output per unit energy consumption. The screw conveying capacity coupling term links the extrusion efficiency target with the degree of matching of conveying capacity. When the conveying capacity is stronger, the larger the extrusion efficiency evaluation index, the better the plasticizing and extrusion coordination. This equation reflects the production law that the faster the feeding speed, the higher the screw speed, the lower the die pressure, the lower the energy consumption rate, and the stronger the screw conveying capacity, the better the extrusion efficiency. By maximizing this objective function, the continuity and stability of production are effectively guaranteed, and the output per unit time is increased by 10% to 15%.

[0097] The improved alternating gray wolf algorithm's position update formula achieves population evolution by weighted averaging of the positions of the three leader wolves and combining the search step size controlled by an adaptive convergence factor. The distance vector describes the positional difference between the current individual and the leader wolf. The adaptive convergence factor dynamically adjusts the search strategy according to population diversity. The average vector is used to evaluate the dispersion of the population's position distribution. When the diversity is high, the search range is increased to avoid premature convergence. When the diversity decreases, the search step size is reduced to accelerate local optimization. This improves the solution quality and accelerates the convergence speed of the algorithm in multi-objective coupled optimization problems.

[0098] In the temperature-controlled flow synergistic balance algorithm, the temperature gradient adjustment formula achieves processing stability control through the coupling calculation of the material temperature change rate and extrusion pressure deviation. This formula considers the rheological mechanism that the viscosity decreases due to the increase in temperature, which in turn causes the pressure to drop. By reducing the temperature gradient amplitude, the trend of continuous temperature increase is suppressed. The screw speed compensation formula compensates for the change in flow resistance caused by the pressure drop by increasing the speed and enhancing the shear effect. At the same time, a residence time constraint factor is introduced to prevent excessive thermal degradation of wood flour. The synergistic adjustment of the two formulas enables the composite material to avoid carbonization and discoloration of heat-sensitive wood flour while ensuring plasticization quality, thus achieving a dynamic balance between thermal stability and flowability.

[0099] The entire dual-game model achieves true coupled optimization of multiple objectives by directly integrating coupling terms into the four objective functions. The first upper-level model uses the temperature field distribution uniformity coupling term... Coupled with wood flour thermal stability margin term The carbonization control is correlated with temperature uniformity and thermal stability margin. The first lower-level model uses a coupling term based on the temperature field distribution uniformity. Coupled with screw conveyor capacity The plasticization uniformity is correlated with temperature uniformity and conveying capacity. The second upper-level model uses an energy consumption rate coupling term. Coupled with wood flour thermal stability margin term The second lower-level model links viscosity control with energy economy and thermal stability margin through an energy consumption rate coupling term. Coupled with screw conveyor capacity By linking extrusion efficiency with energy economy and conveying capacity, the four objective functions achieve global coordination within and across subgames through shared coupling terms. Under high filling conditions of 65% to 75% wood flour, a dynamic balance between thermal stability and flowability is achieved, significantly improving the appearance quality, mechanical properties, processing energy consumption, and production stability of the composite material. This solves the technical problem that traditional single-objective optimization methods cannot simultaneously address the heat sensitivity and high viscosity flowability of wood flour. In one specific embodiment, mold wear is significantly reduced and service life is extended by more than 30%.

[0100] To better understand and implement this invention, Example 2, a specific application scenario, is provided below: In a bio-based composite material processing task, wood flour and polyethylene resin need to be compounded to prepare high-performance granules for subsequent injection molding. The main technical challenge of this task is to ensure that the wood flour does not undergo thermal degradation and carbonization, while also ensuring that the melt has good flowability for stable extrusion, given a wood flour content of 68%. The technical team decided to use the twin-screw compounding and granulation method of this invention to solve this problem.

[0101] In the raw material pretreatment stage, the technical team first sieved the poplar wood powder, obtaining wood powder with a uniform particle size distribution through a 150-mesh standard sieve. Laser particle size analyzer measurements showed that the average particle size was 106 μm, and the particle size distribution conformed to a normal distribution with a standard deviation of 18 μm. The sieved wood powder was then sent to a hot air circulating oven, where the drying temperature was set at 85℃ for 5 hours, reducing the moisture content of the wood powder to 1.6%. High-density polyethylene with a melt flow rate of 10 g / 10 min was selected as the polyethylene resin and mixed with zinc stearate external lubricant at a mass ratio of 100:3. This mixed resin system was used as the matrix material.

[0102] In the gradient temperature-controlled mixing stage, the technical team fed the treated wood flour and mixed polyethylene resin system into a twin-screw extruder with a length-to-diameter ratio of 44, achieving a wood flour mass fraction of 68%. The extruder's feeding section temperature was set at 110℃, the mixing section at 150℃, and the die head section at 160℃, creating a progressively increasing temperature gradient. The screw speed was initially set at 250 r / min, and the feeding rate at 30 kg / h. The twin-screw extruder was equipped with a comprehensive online monitoring system, with multiple temperature sensors installed in the feeding, mixing, and die head sections, collecting data at a frequency of 3 Hz to monitor the material temperature distribution in real time. A pressure sensor was installed at the extruder outlet to measure the extrusion pressure, ranging from 0 MPa to 50 MPa, with a data collection frequency of 3 Hz. A motor power monitoring system collected torque load values ​​at a frequency of 3 Hz.

[0103] like Figure 2 As shown, within the first 15 minutes of processing, the technical team observed that the material temperature distribution gradually increased from 110℃ in the feeding section to 150℃ in the mixing section and then to 160℃ in the die head section, exhibiting a stable gradient. The extrusion pressure gradually decreased from an initial 28MPa to 23MPa during this stage, indicating improved melt flowability. The residence time distribution was measured by adding iron oxide red tracer to the material. A spectrometer at the extruder outlet continuously monitored the tracer concentration change, calculating a peak residence time of 156s and a standard deviation of 24s, indicating relatively stable residence time distribution. When the monitoring system simultaneously detected an upward trend in material temperature distribution, a downward trend in extrusion pressure, and stable residence time distribution, the temperature-controlled flow synergistic balance algorithm automatically activated.

[0104] The temperature-controlled flow synergistic balance algorithm first establishes a dynamic response relationship between material temperature distribution and extrusion pressure. At the 20-minute mark of processing, the temperature sensor in the mixing section detects a local temperature rise to 154℃. The algorithm, recognizing the significant temperature increase, automatically reduces the temperature gradient between the feeding and mixing sections from 40℃ to 35℃, effectively raising the feeding section temperature to 115℃. At this point, the extrusion pressure further decreases to 21MPa, and the algorithm simultaneously increases the screw speed to 265r / min to compensate for changes in flow resistance. Using residence time distribution as a constraint, the algorithm limits the residence time in the high-temperature zone to no more than 180s. By adjusting the feeding rate to 32kg / h, the residence time in the high-temperature zone is controlled at 168s, effectively preventing excessive thermal degradation of the wood flour. The synergistic adjustment of these three parameters ensures that the composite material avoids carbonization and discoloration of the heat-sensitive wood flour while maintaining plasticization quality.

[0105] like Figure 3As shown, the technical team began optimizing processing parameters based on a dual-game model. The first sub-game model targets thermal stability control. The objective function inputs of the first upper-level model include a mixing section temperature of 150℃, a high-temperature zone residence time of 168s, and a wood flour particle size distribution of 106μm. The calculated carbonization degree evaluation index is 11.2, which is below the threshold requirement of 15. The objective function inputs of the first lower-level model include a screw speed of 265r / min, a length-to-diameter ratio of 44, and a material filling rate of 48%. The calculated plasticization uniformity evaluation index is 0.87, which meets the normalization standard of greater than or equal to 0.8. The two models achieve synergistic optimization by using temperature field distribution uniformity as a coupling term. The measured temperature standard deviation in the three regions of feeding section, mixing section, and die head section is 8.3℃, which meets the constraint condition of less than or equal to 10℃.

[0106] The second subgame model focuses on improving fluidity. The objective function inputs of the upper-level model include a wood flour mass fraction of 68%, a zinc stearate external lubricant addition of 3%, and a melt temperature of 158℃. The calculated viscosity increase evaluation index is 276, below the threshold requirement of 300. The objective function inputs of the lower-level model include a feeding rate of 32 kg / h, a screw speed of 265 r / min, and a die pressure of 18 MPa. The calculated extrusion efficiency evaluation index is 0.73, meeting the normalization standard of greater than or equal to 0.6. The two models achieve synergistic optimization through energy consumption rate as a coupling term. The measured energy consumption per unit output is 1.68 kWh / kg, meeting the constraint condition of less than or equal to 2 kWh / kg.

[0107] The first upper-level model and the second upper-level model are coupled through the thermal stability margin of the wood flour. Thermogravimetric analysis determined the critical temperature at which poplar wood flour begins to carbonize to be 178℃, while the actual mixing temperature was 150℃. The calculated thermal stability margin of the wood flour was 28℃, satisfying the constraint condition of ≥20℃. The first lower-level model and the second lower-level model are coupled through the screw conveying capacity. Continuous weighing at the extruder outlet measured the actual conveying capacity to be 31.8 kg / h, while the theoretical conveying capacity, calculated based on the screw geometry, was 36.5 kg / h. The ratio of the actual to the theoretical conveying capacity was 0.87, satisfying the constraint condition of ≥0.85.

[0108] The technical team employed an improved alternating gray wolf algorithm to iteratively solve for the optimal processing parameters. During the parameter population initialization phase, screw speed, feeding rate, mixing section temperature, and zinc stearate external lubricant addition were used as decision variables. Forty initial solutions were randomly generated within their respective process ranges as gray wolf individuals. Each gray wolf individual corresponds to a complete set of processing parameters, as shown in Table 1.

[0109] Table 1. Example of initial parameters for individual gray wolves

[0110] The first iteration optimizes the first subgame model, fixing the zinc stearate external lubricant addition at 3% and the feeding rate at 32 kg / h. The grey wolf algorithm is used to update the mixing section temperature and screw speed. The grey wolf with the best fitness value in the first upper-level model is designated as the alpha wolf (A), corresponding to a mixing section temperature of 148℃, a screw speed of 258 r / min, and a carbonization degree evaluation index of 10.6. The grey wolf with the best fitness value in the first lower-level model is designated as the second-best wolf (A), corresponding to a mixing section temperature of 151℃, a screw speed of 272 r / min, and a plasticization uniformity evaluation index of 0.91. The remaining grey wolves update their positions based on their positional relationships with the alpha, second-best, and last alpha wolves. The initial convergence factor is 2, decreasing linearly with the number of iterations. After 30 iterations in the first round, the current optimal solution for the first subgame model is obtained, with an optimal mixing section temperature of 149℃ and an optimal screw speed of 264 r / min.

[0111] The second iteration optimized the second subgame model, fixing the mixing section temperature at 149℃ and the screw speed at 264 r / min. The gray wolf algorithm was used to update the amount of zinc stearate external lubricant added and the feeding rate. The gray wolf with the best fitness value in the upper-level model was designated as "Head Wolf B," with a corresponding zinc stearate external lubricant addition of 3.2%, a feeding rate of 33 kg / h, and a viscosity increase evaluation index of 268. The gray wolf with the best fitness value in the lower-level model was designated as "Second Wolf B," with a corresponding zinc stearate external lubricant addition of 3.4%, a feeding rate of 35 kg / h, and an extrusion efficiency evaluation index of 0.78. After 30 iterations in the second round, the current optimal solution for the second subgame model was obtained: the optimal zinc stearate external lubricant addition was 3.3%, and the optimal feeding rate was 34 kg / h.

[0112] Two iterations constitute a complete alternating optimization cycle, with the technical team setting a maximum of 150 iterations. During the 112th alternating optimization cycle, the improvement in the objective function value of the first subgame model was less than 0.1% for five consecutive cycles, and the improvement in the objective function value of the second subgame model was also less than 0.1%, satisfying the convergence criterion. The algorithm outputs the final optimal screw speed matching value as 262 r / min, the optimal feeding speed matching value as 34 kg / h, the corresponding optimal mixing section temperature as 149℃, and the optimal zinc stearate external lubricant addition value as 3.3%.

[0113] The improved alternating gray wolf algorithm employs an adaptive nonlinear adjustment mode during the optimization process. The technical team evaluates the diversity of the gray wolf population every 10 iterations, obtaining the diversity index by calculating the standard deviation of the positions of all individual gray wolves. At the 40th iteration, the standard deviation of gray wolf population diversity is 0.38, indicating high diversity. The algorithm increases the convergence factor from the current 1.2 to 1.8 times (2.16) to maintain the exploration range and avoid premature convergence. At the 90th iteration, the standard deviation of gray wolf population diversity decreases to 0.08, indicating decreased diversity. The algorithm decreases the convergence factor from the current 0.6 to 0.5 times (0.3) to accelerate the convergence process. This adaptive adjustment mechanism significantly improves the solution quality of the algorithm in multi-objective coupled optimization problems.

[0114] like Figure 4 As shown, based on the optimized parameters, the technical team adjusted the operating conditions of the twin-screw extruder. The mixing section temperature was adjusted from the initial 150℃ to 149℃, the screw speed from 265 r / min to 262 r / min, the feeding rate from 32 kg / h to 34 kg / h, and the amount of zinc stearate external lubricant added from 3% to 3.3%. The adjusted processing ran continuously for 2 hours, during which the degree of wood flour carbonization and the increase in melt viscosity were monitored in real time. A colorimeter measured the Euclidean distance between the composite material sample and the standard white board in the Lab color space, and the calculated color difference value was 8.6, corresponding to a wood flour carbonization degree evaluation index of 9.8, which is far below the thermal stability threshold of 15, and there is no need to trigger the temperature reduction mechanism. A capillary rheometer was used at a shear rate of 500... Under the conditions, the apparent viscosity of the composite melt was measured to be 1680 Pa·s, and the apparent viscosity of the pure polyethylene matrix was 620 Pa·s. The calculated melt viscosity increase was 171%, and the corresponding viscosity increase evaluation index was 171, which is far below the fluidity threshold of 300. Therefore, there is no need to trigger the lubricant increment mechanism and the speed increase mechanism.

[0115] like Figure 5 As shown, the extrudate passes through a flow channel with a cross-sectional area of ​​360. Wide-channel die forming, with a channel width of 18mm and a channel height of 20mm, compared to the conventional die channel cross-sectional area of ​​250mm. The flow rate increased by 44%. The wide flow channel design significantly reduced the flow resistance of the melt within the die head. The measured die head inlet pressure was 16 MPa, and the die head outlet pressure was 12 MPa, with a pressure drop of only 4 MPa. Melt shear heat generation was effectively controlled, resulting in a uniform temperature distribution within the die head and preventing localized overheating caused by material stagnation due to flow obstruction. The extruded strip material entered a water cooling system with a cooling water temperature of 25°C. The material remained in the water tank for 18 seconds, rapidly cooling the material surface temperature from 160°C to 38°C. The cooled strip material was then cut by a pelletizer with the blade speed set at 120 r / min, yielding cylindrical particles with a length of 3 mm and a diameter of 3 mm.

[0116] After being sieved using a vibrating screen, the granules underwent quality testing. The technical team selected 100 granule samples for visual inspection; the granule surfaces were smooth and uniform, exhibiting a pale yellow color with no obvious carbonized black spots. Scanning electron microscopy revealed the compositional distribution of the granule cross-sections; the wood flour granules were uniformly dispersed within the polyethylene matrix, with a particle spacing standard deviation of 4.2 μm and a plasticization uniformity evaluation index of 0.89. Tensile testing determined the mechanical properties of the composite material granules: tensile strength of 32 MPa, flexural modulus of 3200 MPa, and impact strength of 8.5 kJ / m². All performance indicators met the requirements for subsequent injection molding. The technical team recorded the energy consumption data throughout the entire processing. The average motor power of the twin-screw extruder was 58kW, the cumulative processing time was 2 hours, the total output was 68kg, and the energy consumption per unit output was 1.71kWh / kg. The extrusion efficiency evaluation index, calculated as the ratio of output to energy consumption, was 0.76, indicating good processing economy.

[0117] The technical team conducted a detailed analysis of the temperature field distribution uniformity during the processing. Five temperature monitoring points were set up in the feeding section, with temperatures of 115℃, 116℃, 114℃, 117℃, and 115℃, and a standard deviation of 1.1℃. Six temperature monitoring points were set up in the mixing section, with temperatures of 148℃, 150℃, 149℃, 151℃, 148℃, and 149℃, and a standard deviation of 1.2℃. Five temperature monitoring points were set up in the die head section, with temperatures of 159℃, 161℃, 160℃, 158℃, and 160℃, and a standard deviation of 1.0℃. The average standard deviation of the temperature in the three regions was 7.5℃, meeting the constraint condition of less than or equal to 10℃. The uniform temperature distribution ensured good coordination between the carbonization control target and the plasticization uniformity target during the optimization process.

[0118] The technical team also tested the screw conveying capacity. With a screw speed of 262 r / min, continuous weighing at the extruder outlet for 5 minutes yielded an actual conveying capacity of 33.96 kg / h. Based on the screw geometry (outer diameter 50 mm, screw channel depth 8 mm, pitch 50 mm, length-to-diameter ratio 44), the theoretical conveying capacity was calculated to be 39.2 kg / h. The ratio of actual to theoretical conveying capacity was 0.87, satisfying the constraint condition of ≥0.85. This demonstrates that the screw conveying capacity achieves synergistic optimization of plasticization uniformity and efficiency improvement at a global level.

[0119] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for twin-screw compounding and granulation of bio-based composite materials, characterized in that, The process includes the following steps: raw material pretreatment stage, gradient temperature-controlled mixing stage, real-time monitoring and data collection of wood flour thermal degradation parameters and melt flow resistance parameters, optimization of processing parameters based on a dual game model, and adjustment of the twin-screw extruder operating conditions according to the optimized parameters. Next, the extrudate is formed by a wide-channel die and then cooled by a water-cooling system before being pelletized and packaged. By constructing a first sub-game model for thermal stability control and a second sub-game model for improving flowability, the two sub-game models achieve global coordination through the coupling term of wood flour thermal stability margin and screw conveying capacity. An improved alternating gray wolf algorithm is used to iteratively solve for the optimal screw speed matching value and the optimal feeding speed matching value. When the material temperature distribution value shows an upward trend, the extrusion pressure value shows a downward trend, and the residence time distribution value remains relatively stable, the temperature-controlled flow collaborative balance algorithm is activated to dynamically adjust the temperature gradient and screw speed.

2. The method according to claim 1, characterized in that, The raw material pretreatment stage specifically involves sieving the wood flour to a particle size range of 100-200 mesh, and then placing it in a drying equipment to dry it at a temperature of 80-90°C for 4-6 hours to reduce the moisture content to below 2%.

3. The method according to claim 2, characterized in that, The raw material pretreatment stage also includes screening polyethylene resin according to the melt flow rate standard of 5g / 10min to 15g / 10min and mixing it with zinc stearate external lubricant at a mass ratio of 100:2 to 100:4 for later use.

4. The method according to claim 3, characterized in that, The gradient temperature-controlled mixing stage specifically involves feeding the treated wood flour and polyethylene resin into a twin-screw extruder at a mass ratio of 65:35 to 75:

25. The feeding section temperature is set to 100℃ to 120℃, the mixing section temperature to 140℃ to 160℃, the die head section temperature to 150℃ to 170℃, the screw speed to 200 r / min to 300 r / min, and the length-to-diameter ratio to 40 to 48.

5. The method according to claim 4, characterized in that, The wood flour thermal degradation parameters include material temperature distribution and residence time distribution, and the melt flow resistance parameters include extrusion pressure and torque load.

6. The method according to claim 5, characterized in that, The material temperature distribution value is obtained in real time by installing temperature sensors in the feeding section, mixing section and die head section of the twin-screw extruder, with a sampling frequency of 1Hz to 5Hz.

7. The method according to claim 6, characterized in that, The residence time distribution value is calculated by adding a tracer to the material and detecting the tracer concentration change curve at the extruder outlet. The tracer is an inorganic pigment with good thermal stability.

8. The method according to claim 7, characterized in that, The temperature-controlled flow collaborative balance algorithm achieves processing stability control by constructing a dynamic response relationship between material temperature distribution value and extrusion pressure value. When the material temperature distribution value is detected to be continuously rising, the temperature-controlled flow collaborative balance algorithm automatically reduces the temperature gradient amplitude between the feeding section and the mixing section.

9. The method according to claim 8, characterized in that, In the temperature-controlled flow collaborative balance algorithm, when the extrusion pressure value decreases, the temperature-controlled flow collaborative balance algorithm simultaneously increases the screw speed to compensate for the change in flow resistance. The residence time distribution value is used as a constraint condition to limit the temperature adjustment range to prevent excessive thermal degradation of wood flour.

10. The method according to claim 9, characterized in that, The improved alternating gray wolf algorithm solution process includes an initialization parameter population stage and an alternating iterative solution stage. In the initialization parameter population stage, the screw speed, feeding speed, mixing section temperature, and zinc stearate external lubricant addition amount are used as decision variables.

Images