Injection molding method for producing 6G antenna hole parts

Abstract

The present application provides an injection molding and stamping molding method for producing 6G antenna hole components, including: according to the structural characteristics of the 6G antenna hole components, using computer-aided design software to perform three-dimensional modeling of the parts, obtaining the geometric parameters and hole distribution information of the parts, using the obtained geometric model, and numerically simulating the injection molding process through the finite element analysis method to obtain the temperature field and flow field distribution law in the cavity; in order to solve the problem of uneven temperature field distribution shown in the simulation results, multi-point temperature control technology is used, and multiple temperature sensors and heaters are arranged at key positions of the mold cavity, and the temperature of each area is adjusted in real time through a closed-loop feedback control algorithm to ensure uniform temperature distribution in the cavity and eliminate local overheating or underheating.

Description

Technical Field

[0001] The present invention relates to the technical field of injection molds, and in particular to an injection molding and stamping method for producing 6G antenna hole parts. Background Art

[0002] During the injection molding and stamping process of producing 6G antenna hole components, due to the complex structure of the parts, small hole size and dense distribution, extremely high requirements are placed on the uniformity and stability of the mold temperature field.

[0003] During the injection molding and stamping process, the uneven temperature distribution in the mold cavity will lead to unbalanced melt flow, which is easy to form accumulation or defects around the holes, seriously affecting the dimensional accuracy and surface quality of the parts; in addition, the temperature fluctuation in the cavity will also cause changes in the melt viscosity, resulting in unstable part molding cycle and large batch differences; at the same time, the flow behavior of the melt in the molding process of 6G antenna hole parts is complex and changeable. Under the comprehensive influence of the cavity structure, gating system design and process parameters, the melt filling process in the cavity has uneven flow resistance distribution, which is easy to form defects such as curling and shrinkage at the hole. The uncontrollable melt flow behavior will also lead to uneven residual stress distribution inside the parts, resulting in warping and deformation, and further affecting the dimensional accuracy of the parts. Therefore, it is urgent to establish an accurate multi-physics field coupling simulation model to reveal the key influencing mechanism in the injection molding and stamping process of 6G antenna hole parts, optimize mold design and process control, and ensure stable and reliable part molding quality. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide an injection molding and stamping molding method for producing 6G antenna hole parts, which can improve the molding quality of 6G antenna hole parts.

[0005] In order to solve the above technical problems, the technical solution of the present invention is: an injection molding and stamping molding method for producing 6G antenna hole parts, comprising the following steps:

[0006] S1. According to the structural characteristics of the 6G antenna with holes, the computer-aided design software is used to perform three-dimensional modeling of the parts, and the geometric parameters and hole distribution information of the parts are obtained. The obtained geometric model is used to perform numerical simulation of the injection molding process through the finite element analysis method to obtain the temperature field and flow field distribution law in the cavity; S2. In view of the uneven temperature field distribution problem shown in the simulation results, multi-point temperature control technology is used to arrange multiple temperature sensors and heaters at key positions of the mold cavity. The temperature of each area is adjusted in real time through a closed-loop feedback control algorithm to ensure uniform temperature distribution in the cavity and eliminate local overheating or underheating; S3. Based on the optimized The temperature field distribution is used to optimize the gating system and runner design to improve the melt flow balance. The temperature field distribution data is used to determine the area with uneven heat distribution in the cavity, and the layout of the gating system and runner is adjusted accordingly. The melt filling process under different gating schemes is analyzed through computer fluid dynamics simulation to obtain the optimal gating position and runner cross-sectional size, so that the flow resistance distribution of the melt in the cavity tends to be uniform, avoiding accumulation or defects around the holes; S4. In view of the influence of the temperature and flow field distribution in the cavity on the melt viscosity, an intelligent injection molding process parameter optimization method is adopted, and a temperature-viscosity-pressure mathematical model is established according to the rheological properties of the part material. The model adopts a polynomial fitting method, with temperature and pressure as independent variables and viscosity as the dependent variable. The specific functional relationship is obtained by fitting experimental data. The molding quality under different process parameter combinations is predicted and evaluated through machine learning algorithms to obtain the optimal injection molding temperature, pressure and time parameters, thereby achieving stable control of the part molding cycle. S5. In order to reduce the difference between batches, various sensors such as pressure, temperature, and displacement are installed on the injection molding machine, and online quality monitoring and intelligent feedback control technology are introduced to collect key parameter information of the molding process in real time. The production data of different batches are mined and compared through big data analysis methods to timely discover and correct process deviations and ensure zero The consistency of part quality is ensured. At the same time, by analyzing the batch difference data, the key factors affecting the residual stress distribution are identified to provide a basis for the subsequent residual stress control; S6. In view of the warping deformation problem caused by the uneven distribution of residual stress inside the parts, based on the aforementioned optimized process parameters and batch difference analysis results, a multi-step injection molding process is adopted. The process includes the following steps: first, pre-filling is performed, then segmented filling is performed according to the preset time interval and pressure gradient, and finally, pressure holding and cooling are performed. Through this layered filling and sequential cooling method, the flow and solidification process of the melt in the cavity is controlled, so that the residual stress distribution inside the part is more uniform and the deformation risk is reduced;S7. To further improve the dimensional accuracy and surface quality of parts, a precision finishing process is introduced after injection molding and stamping. Based on the temperature field distribution, flow characteristics and residual stress distribution data obtained in the previous steps, the areas where defects may occur on the surface of the parts are determined. For these areas, ultrasonic vibration or laser drilling and other technologies are used to remove and repair the tiny defects on the surface of the parts. The key dimensions of the parts are measured online and feedback controlled through the machine vision inspection system, and the finishing process parameters are continuously optimized to finally meet the high-precision requirements of 6G antenna components. ;

[0007] Furthermore, the step S1 specifically includes:

[0008] According to the structural characteristics of the 6G antenna hole parts, the parts are 3D modeled to obtain the geometric parameters and hole distribution information of the parts;

[0009] Based on the obtained geometric model and hole distribution information, a finite element analysis model of the injection molding process is constructed;

[0010] According to the constructed finite element analysis model, the process parameters of the injection molding process are set, including injection temperature, injection pressure and stamping speed;

[0011] According to the set process parameters, the numerical simulation calculation of the injection molding process is carried out to obtain the temperature field and flow field distribution law in the cavity;

[0012] According to the temperature field distribution law obtained by numerical simulation, determine whether the temperature distribution on the surface of the part is uniform during the injection molding process. If it is not uniform, adjust the injection temperature and cooling time;

[0013] According to the flow field distribution law obtained by numerical simulation, it is judged whether the melt filling is smooth during the injection molding process. If not, the injection pressure and runner design are adjusted;

[0014] According to the adjusted process parameters, the numerical simulation calculation of the injection molding process is carried out again until the temperature field and flow field distribution that meet the requirements are obtained, and the injection molding process optimization of the 6G antenna hole parts is completed.

[0015] Further, the step S2 specifically includes:

[0016] According to the structural characteristics of the mold cavity and the temperature distribution requirements, the key positions are determined and temperature sensors and heaters are arranged at the key positions. The temperature sensors are used to collect temperature data of each area in real time, and the heaters are used for local heating.

[0017] After the temperature sensor collects the real-time temperature data of each area, it transmits the temperature data to the closed-loop control system. The closed-loop control system determines whether the temperature of each area meets the requirements based on the preset target temperature distribution;

[0018] If the temperature of a certain area is lower than the target temperature, the closed-loop control system sends a heating instruction to the heater in that area to increase the temperature of that area;

[0019] If the temperature of a certain area is higher than the target temperature, the closed-loop control system sends a stop heating or cooling instruction to the heater in that area to reduce the temperature of that area;

[0020] The closed-loop control system uses the PID control algorithm to calculate the heating power and heating time of each heater based on the real-time temperature data and target temperature distribution of each area, generates accurate temperature control instructions, and realizes accurate adjustment of the temperature of each area;

[0021] During the temperature control process, the closed-loop control system also needs to dynamically adjust the parameters of the PID control algorithm based on the real-time temperature data fed back by the temperature sensor to adapt to the changes in the heat conduction characteristics inside the mold cavity and improve the accuracy and response speed of temperature control;

[0022] The closed-loop control system uses an adaptive control algorithm to optimize the temperature control strategy according to the temperature status and control effect of each area, automatically adjusts the target temperature and control parameters of each area, and improves the efficiency of temperature control and energy utilization while meeting the temperature uniformity requirements;

[0023] During the temperature control process, the closed-loop control system monitors the temperature change trend of each area in real time, and uses a predictive control algorithm to predict the risk of temperature fluctuations in advance. When it is predicted that a certain area may have abnormal temperature, the temperature of the adjacent area is adjusted in advance to prevent local overheating or underheating, ensuring that the temperature in the mold cavity is always in a uniform and stable state.

[0024] Further, the step S3 specifically includes:

[0025] According to the optimized temperature field distribution data, determine the area with uneven heat distribution in the cavity and determine the area that needs to be optimized;

[0026] Computer fluid dynamics simulation technology is used to simulate the melt filling process of different gating systems and runner layout schemes to obtain the flow resistance distribution under each scheme;

[0027] By comparing the simulation results of different schemes, the best combination of pouring position and runner cross-section size is determined to make the flow resistance of the melt in the cavity most evenly distributed;

[0028] If the simulation results show that the flow resistance in certain areas is still large, the gating system and runner layout of the area should be adjusted accordingly, and the simulation verification should be repeated;

[0029] According to the gating system and runner layout scheme obtained through simulation optimization, the actual gating process parameters are set, including the gating temperature and gating speed;

[0030] During the trial production process, the quality of the finished castings is tested, with a focus on whether there are defects near the area of ​​uneven heat distribution in the cavity, and relevant data is collected;

[0031] Based on the feedback from the trial production data, the gating system and runner layout are further optimized and adjusted until the melt flow balance and casting product quality requirements are met.

[0032] Further, the step S4 specifically includes:

[0033] According to the real-time data of the injection molding process, the temperature, pressure and viscosity parameters of the melt are obtained, and a three-dimensional mathematical model of temperature-viscosity-pressure is established;

[0034] The polynomial fitting method is adopted, with temperature and pressure as independent variables and viscosity as dependent variable, and a large amount of experimental data is fitted to obtain the specific functional relationship between temperature, pressure and viscosity;

[0035] The fitted mathematical model is used as the input of the machine learning algorithm to train the support vector machine, random forest and neural network models to predict and evaluate the molding quality under different process parameter combinations;

[0036] According to the prediction results of the machine learning model, the genetic algorithm optimization method is used to search for the optimal combination of injection temperature, pressure and time parameters to obtain an intelligent injection molding process parameter optimization solution;

[0037] The optimized process parameters are input into the injection molding machine control system. Through closed-loop feedback control, the temperature, pressure and time parameters of the injection molding process are adjusted in real time to ensure that the melt viscosity is within the optimal range.

[0038] During the injection molding process, real-time data on temperature, pressure and viscosity are continuously collected to update and optimize mathematical models and machine learning algorithms online, continuously improving the accuracy and stability of process parameter optimization;

[0039] By integrating with the injection molding machine control system, the whole process of injection molding can be intelligently controlled, the part molding cycle can be shortened, and product quality and production efficiency can be improved.

[0040] Further, the step S5 specifically includes:

[0041] According to the requirements of the injection molding process, determine the key parameters of pressure, temperature and displacement that need to be monitored, select sensors of appropriate models and specifications, and install them at key locations of the injection molding machine;

[0042] The parameter information collected by the sensor is acquired in real time through the data acquisition module, and the data is preprocessed, including denoising and normalization, to obtain standardized parameter data;

[0043] The standardized parameter data is stored in the big data platform, and the production data of different batches are compared and analyzed using time series analysis and cluster analysis data mining algorithms to identify the normal range and abnormal deviation of each parameter;

[0044] If an abnormal deviation is found in a batch of parameters, an early warning is triggered, and the expert knowledge base is used to match the possible causes and give corrective suggestions to guide operators to adjust process parameters in a timely manner;

[0045] Conduct statistical analysis on the difference data between batches, calculate the correlation between each parameter and the residual stress distribution of parts, identify the key factors that have a greater impact on residual stress, and form a factor weight list;

[0046] According to the factor weight list, a residual stress distribution prediction model is established, and the parameter data collected in real time is input to predict the residual stress distribution of the molded parts. When the prediction result exceeds the allowable range, an early warning is triggered and optimization suggestions are given;

[0047] The optimized process parameters are written into the injection molding machine control system, and parameter adaptive optimization is achieved through closed-loop feedback control, which continuously reduces batch differences, improves part quality consistency, and continuously accumulates an optimization knowledge base.

[0048] Further, the step S6 specifically includes:

[0049] According to the structural characteristics and material properties of the parts, a numerical simulation model of the injection molding process is established, and the optimal combination of injection molding process parameters, including injection temperature, injection speed, holding pressure and time, is determined through simulation analysis;

[0050] The injection molding process is monitored in real time using machine vision technology. The position of the melt flow front and the cavity filling status are identified through image processing algorithms. When uneven filling is detected, the injection speed and pressure distribution are dynamically adjusted to achieve optimal control of the melt filling process.

[0051] Obtain the internal temperature distribution data of the part, combine it with the material's pvT characteristic curve, calculate the residual stress distribution through finite element analysis, and adjust the injection molding process parameters when the stress distribution unevenness exceeds the preset threshold;

[0052] Establish a data-driven model for injection molding product quality and process parameters, use machine learning algorithms such as support vector machines or neural networks, obtain a quality prediction model through training with a large amount of historical data, optimize the injection molding process settings based on the model prediction results, and improve the product qualification rate;

[0053] Real-time collection and statistical analysis of the pressure and temperature parameters of the injection molding process are carried out to build a multivariate linear regression model, explore the quantitative relationship between each parameter and the deformation of the part, determine the key control parameters, and guide the optimization of the injection molding process;

[0054] For different batches of raw materials, infrared spectroscopy nondestructive testing methods are used to characterize their performance parameters. The test data is input into the quality prediction model to dynamically adjust the injection molding process parameters of the batch to reduce the impact of batch differences on product quality.

[0055] In the molding process, vibration or ultrasonic external field is introduced to adjust the frequency and energy parameters of the external field to induce the melt to produce a uniform micro-orientation, improve the residual stress distribution inside the part, and achieve the purpose of reducing deformation.

[0056] Further, the step S7 specifically includes:

[0057] Based on the temperature field distribution, flow characteristics and residual stress distribution data obtained from the injection molding process, a part surface defect prediction model is constructed, and the support vector machine algorithm is used to identify and locate areas where defects may occur;

[0058] For the identified defective area, a high-definition image of the defective area is obtained through a machine vision system, and an image segmentation algorithm is used to extract the precise contour of the defective area;

[0059] According to the morphological characteristics of the defect area, determine whether to use ultrasonic vibration or laser drilling technology to remove and repair the defect;

[0060] If the defect area is small and regular in shape, ultrasonic vibration technology is used;

[0061] If the defect area is large and irregular in shape, laser drilling technology is used;

[0062] During defect removal and repair, the machine vision system is used to obtain real-time images of the part surface, and the image quality assessment algorithm is used to determine the repair effect;

[0063] If the trimming effect does not reach the preset threshold, the process parameters of ultrasonic vibration or laser drilling are adjusted and the trimming process is repeated;

[0064] After trimming is completed, the key dimensions of the parts are measured online through a machine vision system, and the Kalman filter algorithm is used to filter and fuse the measurement results to obtain a high-precision estimate of the part dimensions;

[0065] Compare the estimated part size with the designed size and calculate the dimensional deviation;

[0066] If the deviation exceeds the preset tolerance range, the injection molding and stamping process parameters and precision finishing process parameters are adjusted, and the surface defect identification is returned and re-executed;

[0067] If the part's dimensional deviation and surface quality meet the requirements, the part will be transferred to the next process as a qualified product;

[0068] At the same time, the optimized process parameters are saved in the process database to provide reference for subsequent batch production.

[0069] The beneficial effects of the present invention are as follows: the present invention discloses an injection molding and stamping method for 6G antenna hole parts, which first obtains part geometric parameters through three-dimensional modeling, simulates the injection molding process using finite element analysis, and uses multi-point temperature control technology to adjust the cavity temperature in real time to address the problem of uneven temperature field distribution. Based on the optimized temperature field, the casting system is designed to improve melt flow, and the injection molding process parameters are optimized by establishing a temperature-viscosity-pressure mathematical model and a machine learning algorithm. Online quality monitoring technology is introduced, and batch differences are analyzed to provide a basis for residual stress control. A multi-step injection molding process is used to control the residual stress distribution and reduce the risk of deformation. Finally, a precision finishing process is introduced to improve part accuracy and surface quality. The present invention realizes high-precision intelligent production of 6G antenna parts, effectively solves the problems of uneven temperature field and uneven residual stress distribution, and significantly improves part quality and production efficiency. Description of the drawings

[0070] Figure 1 The present invention is a flow chart of an injection molding and stamping molding method for producing 6G antenna hole parts.

[0071] Figure 2 The present invention is a flow chart of step S3 of an injection molding and stamping molding method for producing 6G antenna hole parts.

[0072] Figure 3 This is a schematic flow chart of step S4 of an injection molding and stamping molding method for producing a 6G antenna hole component of the present invention.

[0073] In order to make the purpose, technical solution and advantages of the present invention more clear, the present invention is further described in detail below in conjunction with the accompanying drawings and specific implementation methods. The following embodiments can enable technicians in this field to understand the present invention more comprehensively, but the present invention is not limited to the scope of the embodiments.

[0074] like Figure 1-3, this specific implementation adopts the following technical solution: an injection molding and stamping molding method for producing 6G antenna hole parts, comprising the following steps:

[0075] S1. According to the structural characteristics of the 6G antenna hole parts, computer-aided design software is used to perform three-dimensional modeling of the parts to obtain the geometric parameters and hole distribution information of the parts. Using the obtained geometric model, the injection molding process is numerically simulated through the finite element analysis method to obtain the temperature field and flow field distribution law in the cavity.

[0076] According to the structural characteristics of the 6G antenna hole parts, the parts are 3D modeled to obtain the geometric parameters and hole distribution information of the parts; based on the obtained geometric model and hole distribution information, a finite element analysis model of the injection molding and stamping molding process is constructed; based on the constructed finite element analysis model, the process parameters of the injection molding and stamping molding process are set, including injection temperature, injection pressure, and stamping speed; according to the set process parameters, the injection molding and stamping molding process is numerically simulated to obtain the temperature field and flow field distribution law in the cavity; according to the temperature field distribution law obtained by numerical simulation calculation, it is judged whether the temperature distribution on the surface of the part is uniform during the injection molding and stamping molding process, if not, the injection temperature and cooling time are adjusted; according to the flow field distribution law obtained by numerical simulation calculation, it is judged whether the melt filling is smooth during the injection molding and stamping molding process, if not, the injection pressure and runner design are adjusted; according to the adjusted process parameters, the numerical simulation calculation of the injection molding and stamping molding process is re-performed until the temperature field and flow field distribution that meet the requirements are obtained.

[0077] Specifically, the 6G antenna hole parts were 3D modeled using SolidWorks software. The geometric dimensions of the parts were obtained by measurement. The hole diameter was 8mm, the spacing was 2mm, and the holes were distributed in a rectangular array. The CAD model was imported into ANSYS and tetrahedral meshing was used. The unit size was 1mm and the number of meshes was about 500,000. ABS plastic was used as the material, with a density of 0.5g / cm³ and an elastic modulus of 25GPa.

[0078] According to manufacturing experience, the initial setting of injection temperature is 230℃, injection pressure is 80MPa, and punching speed is 5mm / s. Through numerical simulation, it is found that the surface temperature of the part is unevenly distributed, the temperature in the center area is higher, and the maximum temperature difference is 15℃. Therefore, the injection temperature is adjusted to 220℃, and the cooling time is extended by 2s. At the same time, it is found that the filling near the hole is not smooth, and there is a risk of underfilling. After analysis, it is found that the hole hinders the flow of the melt, so the injection pressure is increased to 100MPa, and the gate and runner design are optimized to make the melt flow around. After 3 iterative calculations, qualified temperature field and flow field results are finally obtained, the temperature difference is controlled within 5℃, and the filling is full and defect-free.

[0079] S2. In order to solve the problem of uneven temperature field distribution shown in the simulation results, multi-point temperature control technology is adopted. Multiple temperature sensors and heaters are arranged at key positions of the mold cavity. The temperature of each area is adjusted in real time through a closed-loop feedback control algorithm to ensure uniform temperature distribution in the cavity and eliminate local overheating or underheating.

[0080] According to the structural characteristics of the mold cavity and the temperature distribution requirements, the key positions are determined, and temperature sensors and heaters are arranged at the key positions. The temperature sensors are used to collect the temperature data of each area in real time, and the heaters are used for local heating. After the temperature sensor collects the real-time temperature data of each area, the temperature data is transmitted to the closed-loop control system. The closed-loop control system determines whether the temperature of each area meets the requirements based on the preset target temperature distribution. If the temperature of a certain area is lower than the target temperature, the closed-loop control system sends a heating instruction to the heater of the area to increase the temperature of the area. If the temperature of a certain area is higher than the target temperature, the closed-loop control system sends a stop heating or cooling instruction to the heater of the area to reduce the temperature of the area.

[0081] The closed-loop control system uses the PID control algorithm to calculate the heating power and heating time of each heater according to the real-time temperature data and target temperature distribution of each area, generates accurate temperature control instructions, and realizes accurate adjustment of the temperature of each area. During the temperature control process, the closed-loop control system also needs to dynamically adjust the parameters of the PID control algorithm according to the real-time temperature data fed back by the temperature sensor to adapt to the changes in the heat conduction characteristics inside the mold cavity and improve the accuracy and response speed of temperature control. According to the temperature status and control effect of each area, the closed-loop control system uses an adaptive control algorithm to optimize the temperature control strategy, automatically adjust the target temperature and control parameters of each area, and improve the efficiency and energy utilization of temperature control while meeting the temperature uniformity requirements. During the temperature control process, the closed-loop control system monitors the temperature change trend of each area in real time, and uses a predictive control algorithm to predict the risk of temperature fluctuation in advance. When it is predicted that a certain area may have abnormal temperature, the temperature of the adjacent area is adjusted in advance to prevent local overheating or underheating, and ensure that the temperature in the mold cavity is always in a uniform and stable state.

[0082] Specifically, according to the structural characteristics and temperature distribution requirements of the mold cavity, PT100 platinum resistance temperature sensors and ceramic heaters are arranged at key positions such as the four corners and center of the cavity; the temperature sensor collects temperature data of each area every 1 second and transmits it to the closed-loop control system based on the STM32 single-chip microcomputer. The control system calculates the PWM duty cycle of each heater based on the preset target temperature distribution (such as 180°C for the four corner areas and 200°C for the center area). For example, according to the temperature deviation and PID parameters P=5, I=2, D=1, when the temperature of the corresponding area is lower than the target temperature, the PWM duty cycle is 80%; when it is higher than the target temperature, the PWM duty cycle is 30%; at the end of each temperature sampling cycle of the control system, the PID parameters are dynamically adjusted using the fuzzy adaptive algorithm according to the average temperature deviation within the cycle, and the temperature change trend of the next cycle is predicted based on the Kalman filter algorithm. The heating power of the adjacent areas is adjusted 5 seconds in advance to control the temperature fluctuation amplitude within ±2°C to ensure the uniformity and stability of the temperature distribution in the mold cavity.

[0083] S3. Based on the optimized temperature field distribution, the gating system and runner are designed and optimized to improve the melt flow balance. Specifically, the temperature field distribution data is used to determine the area with uneven heat distribution in the cavity, and the layout of the gating system and runner is adjusted accordingly. The melt filling process under different casting schemes is analyzed through computer fluid dynamics simulation to obtain the optimal casting position and runner cross-sectional size, so that the flow resistance distribution of the melt in the cavity tends to be uniform, avoiding accumulation or defects around the holes.

[0084] According to the optimized temperature field distribution data, the area with uneven heat distribution in the cavity is judged, and the parts that need to be optimized are determined; computer fluid dynamics simulation technology is used to simulate the melt filling process of different casting systems and runner layout schemes to obtain the flow resistance distribution under each scheme; by comparing the simulation results of different schemes, the best casting position and runner cross-sectional size combination is determined to make the flow resistance distribution of the melt in the cavity most uniform; if the simulation results show that the flow resistance in some areas is still large, the casting system and runner layout of the area are adjusted in a targeted manner, and the simulation verification is re-performed; according to the casting system and runner layout scheme obtained by simulation optimization, the actual casting process parameters are set, including casting temperature and casting speed; during the trial production process, the quality of the finished castings is tested, focusing on whether there are defects near the area with uneven heat distribution in the cavity, and collecting relevant data; according to the trial production data feedback, the casting system and runner layout scheme are further optimized and adjusted until the melt flow balance and casting finished product quality requirements are met.

[0085] Specifically, firstly, according to the optimized temperature field distribution data, the finite element analysis software ANSYS was used to analyze the temperature distribution in the cavity. By comparing the temperature gradients of various areas, it was found that the temperature gradients in the upper left and lower right corners of the cavity were large, and the heat distribution was uneven, which needed to be optimized. Then, the CFD software Fluent was used to simulate the melt filling process of different gating systems and runner layout schemes. By solving the Navier-Stokes equation, the flow velocity and pressure distribution cloud maps under each scheme were obtained, and the flow resistance of each area was calculated. By comparing three different combinations of pouring positions and runner cross-sectional dimensions, it was found that one of the schemes had the most uniform flow resistance distribution, with a standard deviation of only 12 MPa. However, the flow resistance in the lower right corner still reached 35 MPa. Targetedly, the gate diameter of this area was increased from the original 2 cm to 3 cm, and the runner cross-sectional area was increased by 20%. The simulation verification was repeated, and the flow resistance was reduced to 29 MPa. According to the optimization plan, the pouring temperature was set to 720℃ and the pouring speed was set to 5m / s. A trial production of 100 castings was carried out, with a focus on testing the lower right corner area. Two castings were found to have shrinkage defects with a defect rate of 2%. After secondary optimization, the pouring temperature was increased to 730℃, the pouring speed was reduced to 2m / s, and the defect rate was reduced to 5%, meeting the requirements of melt flow balance and finished product quality.

[0086] S4. Aiming at the influence of cavity temperature and flow field distribution on melt viscosity, an intelligent injection molding process parameter optimization method is adopted; according to the rheological properties of the part material, a temperature-viscosity-pressure mathematical model is established; the model adopts a polynomial fitting method, with temperature and pressure as independent variables and viscosity as the dependent variable, and obtains a specific functional relationship through fitting experimental data; the molding quality under different process parameter combinations is predicted and evaluated through machine learning algorithms, and the optimal injection molding temperature, pressure and time parameters are obtained to achieve stable control of the part molding cycle.

[0087] According to the real-time data of the injection molding process, the temperature, pressure and viscosity parameters of the melt are obtained, and a three-dimensional mathematical model of temperature-viscosity-pressure is established; the polynomial fitting method is adopted, with temperature and pressure as independent variables and viscosity as dependent variable, and a large amount of experimental data is fitted to obtain the specific functional relationship between temperature, pressure and viscosity; the fitted mathematical model is used as the input of the machine learning algorithm, and models such as support vector machine SVM, random forest and neural network are trained to predict and evaluate the molding quality under different process parameter combinations; according to the prediction results of the machine learning model, optimization methods such as genetic algorithm are used to search for the optimal injection temperature and pressure. The temperature, pressure and time parameters are combined to obtain an intelligent injection molding process parameter optimization solution; the optimized process parameters are input into the injection molding machine control system, and the temperature, pressure, time and other parameters in the injection molding process are adjusted in real time through closed-loop feedback control to ensure that the melt viscosity is within the optimal range; during the injection molding process, real-time data such as temperature, pressure and viscosity are continuously collected, and the mathematical model and machine learning algorithm are updated and optimized online to continuously improve the accuracy and stability of process parameter optimization; through integration with the injection molding machine control system, the full process of injection molding is intelligently controlled, the part molding cycle is shortened, and product quality and production efficiency are improved.

[0088] Specifically, during the injection molding process, the temperature, pressure, viscosity and other parameters of the melt are collected in real time through sensors, where the temperature range is 180℃~220℃, the pressure range is 50MPa~80MPa, and the viscosity range is 100Pa·s~500Pa·s. Using the polynomial fitting method, with temperature and pressure as independent variables and viscosity as the dependent variable, 500 sets of experimental data are fitted to obtain the functional relationship between temperature, pressure and viscosity: η=05T²-8P²+1TP+200, where η is viscosity, T is temperature, and P is pressure. The mathematical model is input into machine learning models such as support vector machine, random forest and BP neural network, and the molding quality data under 5000 sets of different process parameter combinations are used as training sets to train the model, and the average prediction accuracy can reach more than 95%. On this basis, a genetic algorithm is used to search for the optimal combination of process parameters, with molding cycle and product qualification rate as the optimization targets, crossover probability of 8, mutation probability of 1, population size of 100, and the optimal injection temperature of 205℃, pressure of 65MPa, and holding time of 10s after 50 generations of evolution. The optimized process parameters are input into the injection molding machine control system, and the injection molding process parameters are adjusted in real time through the PID closed-loop control algorithm. The control accuracy can reach ±5℃, ±1MPa, and ±1s, ensuring that the melt viscosity is stable at about 300Pa·s; at the same time, real-time data is continuously collected to update the mathematical model and machine learning algorithm online, and an iterative optimization is performed after every 1,000 products are produced to continuously improve the accuracy and stability of process parameter optimization. Through the above intelligent control method, the injection molding cycle can be shortened by 15%, the product qualification rate can be increased by 5%, and the production efficiency can be increased by 20%.

[0089] S5. In order to reduce the difference between batches, various sensors such as pressure, temperature, displacement, etc. are installed on the injection molding machine, and online quality monitoring and intelligent feedback control technology are introduced; key parameter information of the molding process is collected in real time, and the production data of different batches are mined and compared through big data analysis methods, so as to timely discover and correct process deviations and ensure the consistency of part quality; at the same time, by analyzing the difference data between batches, the key factors affecting the residual stress distribution are identified to provide a basis for subsequent residual stress control.

[0090] According to the requirements of the injection molding process, determine the key parameters of pressure, temperature, and displacement that need to be monitored, select sensors of appropriate models and specifications, and install them at key locations of the injection molding machine; obtain parameter information collected by the sensor in real time through the data acquisition module, and pre-process the data, including denoising and normalization, to obtain standardized parameter data; store the standardized parameter data in the big data platform, and use data mining algorithms such as time series analysis and cluster analysis to compare and analyze production data from different batches to identify the normal range and abnormal deviations of each parameter; if an abnormal deviation is found in a batch of parameters, an early warning is triggered, and possible causes are matched through the expert knowledge base, and corrections are given Suggestions are given to guide operators to adjust process parameters in time; statistical analysis is performed on batch difference data, the correlation between each parameter and the residual stress distribution of parts is calculated, the key factors that have a greater impact on residual stress are identified, and a factor weight list is formed; according to the factor weight list, a residual stress distribution prediction model is established, and the parameter data collected in real time is input to predict the residual stress distribution of molded parts. When the prediction result exceeds the allowable range, an early warning is triggered and optimization suggestions are given; the optimized process parameters are written into the injection molding machine control system, and parameter adaptive optimization is achieved through closed-loop feedback control, which continuously reduces batch differences, improves part quality consistency, and continuously accumulates an optimization knowledge base.

[0091] Specifically, according to the requirements of the injection molding process, Kistler 6152A pressure sensor is selected to monitor the injection pressure, with a range of 0-2000bar and an accuracy of ±5%FS; Pt100 temperature sensor is selected to monitor the barrel temperature, with a range of 0-400℃ and an accuracy of ±5℃; KEYENCE IL-300 laser displacement sensor is selected to monitor the injection displacement, with a range of 0-800mm and an accuracy of ±02%FS. The sensors were installed at key positions such as the barrel and mold of the injection molding machine. The data acquisition module based on LabVIEW was used to collect sensor data in real time at a frequency of 1kHz, and the 5th-order Butterworth low-pass filter was used for denoising and maximum and minimum value normalization. The processed data was stored in the Hadoop big data platform. The ARIMA time series model and K-means clustering algorithm were used to compare and analyze different batches of data, and the normal ranges of pressure, temperature, and displacement were identified to be 800-1200bar, 180-220℃, and 300-500mm, respectively. The abnormal deviation threshold was ±5%. If an abnormality was found, an early warning would be triggered. Rule matching was performed through the expert knowledge base based on Drools, and corrective suggestions such as "excessive injection pressure may be due to gate blockage, and it is recommended to check the gate and clean it" were given. The Pearson correlation coefficient method was used to calculate the correlation between each parameter and the residual stress of the part, and it was found that injection pressure, barrel temperature, and holding time were the key factors affecting residual stress, with weights of 6, 3, and 1, respectively. Based on this, a residual stress prediction model based on BP neural network was established. Real-time parameter data was input to predict the residual stress distribution. When the prediction result exceeded 50MPa, an early warning was triggered. The process parameters were optimized through genetic algorithm, and the optimal parameter combination was obtained as injection pressure 1000bar, barrel temperature 200℃, and holding time 10s. The optimal parameter combination was written into the Siemens S7-1500 PLC control system, and the closed-loop control of the injection molding machine parameters was realized through Profinet industrial Ethernet. The optimization knowledge was continuously accumulated in the MongoDB knowledge base to form an intelligent optimization closed loop of the injection molding process.

[0092] S6. To solve the warping problem caused by uneven residual stress distribution inside parts, a multi-step injection molding process is adopted based on the above-mentioned optimized process parameters and batch difference analysis results. The process includes the following steps: first, pre-filling, then segmented filling according to the preset time interval and pressure gradient, and finally pressure holding and cooling; through this layered filling and sequential cooling method, the flow and solidification process of the melt in the cavity is controlled, so that the residual stress distribution inside the part is more uniform and the deformation risk is reduced.

[0093] According to the structural characteristics and material properties of the parts, a numerical simulation model of the injection molding process is established, and the optimal combination of injection molding process parameters, including injection temperature, injection speed, holding pressure and time, is determined through simulation analysis. The injection molding process is monitored in real time by machine vision technology, and the position of the melt flow front and the cavity filling state are identified through image processing algorithms. When it is judged that the filling is uneven, the injection speed and pressure distribution are dynamically adjusted to achieve optimized control of the melt filling process; the internal temperature distribution data of the parts are obtained, and the residual stress distribution is calculated by the finite element analysis method in combination with the pvT characteristic curve of the material. When the unevenness of the stress distribution exceeds the preset threshold, the injection molding process parameters are adjusted, such as extending the cooling time or optimizing the cooling system layout; a data-driven model of injection molding product quality and process parameters is established, and machine learning algorithms such as support vector machines or neural networks are used to obtain a quality prediction model through training with a large amount of historical data. The injection molding process settings are optimized according to the model prediction results to improve the product qualification rate.

[0094] The pressure, temperature and other parameters of the injection molding process are collected and statistically analyzed in real time, and a multivariate linear regression model is constructed to explore the quantitative relationship between each parameter and the deformation of the part, determine the key control parameters, and guide the optimization of the injection molding process. For different batches of raw materials, non-destructive testing methods such as infrared spectroscopy are used to characterize their performance parameters, such as viscosity, melt flow rate, etc. The test data are input into the quality prediction model to dynamically adjust the injection molding process parameters of the batch to reduce the impact of batch differences on product quality. In the molding process, external fields such as vibration or ultrasound are introduced to induce the melt to produce a uniform micro-orientation by adjusting parameters such as the external field frequency and energy, thereby improving the residual stress distribution inside the part and reducing deformation.

[0095] Specifically, according to the structural characteristics and material properties of the parts, Moldflow software was used to establish a three-dimensional numerical simulation model of the injection molding process. Through orthogonal experimental design, factors such as injection temperature of 280-320℃, injection speed of 60-120mm / s, holding pressure of 60-120MPa and holding time of 5-15s were selected. Each factor had 3 levels, with a total of 27 experiments. The simulation analysis obtained the optimal process parameter combination of injection temperature of 300℃, injection speed of 90mm / s, holding pressure of 90MPa and holding time of 10s. A high-speed camera system based on CCD sensor was used to monitor the injection molding process in real time. The image acquisition frequency was 1000 frames / second. The image segmentation algorithm based on the regional growing method was used to extract the position of the melt flow front and calculate the cavity filling rate. When the filling rate deviation exceeded 5%, the injection speed and pressure distribution were dynamically adjusted through the PID control algorithm to achieve optimal control of the melt filling process.

[0096] Four thermocouple sensors are arranged in the cavity to collect the internal temperature distribution data of the part. Combined with the pvT characteristic curve of the material, the finite element analysis software ANSYS is used to divide 100,000 tetrahedral units and calculate the residual stress distribution cloud map. When the stress distribution unevenness exceeds 20%, the injection molding process parameters are optimized by the response surface method, the cooling time is extended to 30s, and the cooling water channel layout is optimized to reduce the cooling water channel spacing by 20% to improve the residual stress distribution. A data-driven model of the dimensional accuracy and process parameters of injection molding products is established. 500 samples are randomly selected, 20 key dimensions of each sample are measured, and 10,000 sets of dimensional data and corresponding process parameter settings are obtained. The support vector machine algorithm is used to train the dimensional accuracy prediction model, and the prediction accuracy is 95% through cross-validation. According to the model prediction results, the injection molding process parameters are optimized to improve the product qualification rate to more than 98%.

[0097] The pressure, temperature and other parameters of the injection molding process are collected in real time. 100 sets of data are collected in each cycle, and a total of 100,000 sets of data are collected to construct a multivariate linear regression model. Through variance analysis, it is determined that pressure and temperature are the key factors affecting the deformation of parts. For every 10MPa increase in pressure, the deformation increases by 5mm, and for every 10℃ increase in temperature, the deformation increases by 3mm. For different batches of PP raw materials, the melt flow rate is determined by infrared spectroscopy, and the viscosity is determined by rotational rheometer. The performance parameter deviation of different batches of materials is within 5%. The test data is input into the quality prediction model to dynamically adjust the injection molding pressure and temperature. The pressure adjustment range is ±5MPa, and the temperature adjustment range is ±10℃ to reduce the impact of batch differences on product quality. During the molding process, 4 piezoelectric ceramic vibrators are arranged on the inner wall of the cavity with a vibration frequency of 20kHz and a vibration energy of 100W. Through finite element analysis, the layout of the vibrators is optimized to make the melt evenly stressed and induce the melt to produce uniform micro-orientation. The unevenness of residual stress distribution is reduced to less than 10%, effectively reducing part deformation.

[0098] S7. To further improve the dimensional accuracy and surface quality of parts, a precision finishing process is introduced after injection molding and stamping. Based on the temperature field distribution, flow characteristics and residual stress distribution data obtained in the previous steps, the areas where defects may occur on the surface of the parts are determined. For these areas, ultrasonic vibration or laser drilling and other technologies are used to remove and repair tiny defects on the surface of the parts. The key dimensions of the parts are measured online and feedback controlled through a machine vision inspection system, and the finishing process parameters are continuously optimized to ultimately meet the high-precision requirements of 6G antenna components.

[0099] Based on the temperature field distribution, flow characteristics and residual stress distribution data obtained from the injection molding process, a part surface defect prediction model is constructed, and the support vector machine algorithm is used to identify and locate areas where defects may occur; for the identified defective areas, a high-definition image of the defective area is obtained through a machine vision system, and an image segmentation algorithm is used to extract the precise contour of the defective area; based on the morphological characteristics of the defective area, it is determined to use ultrasonic vibration or laser drilling technology for defect removal and repair; if the defective area is small and regular in shape, ultrasonic vibration technology is used; if the defective area is large and irregular in shape, laser drilling technology is used; in the process of defect removal and repair, the part surface image is obtained in real time through a machine vision system, and an image quality evaluation algorithm is used to judge the finishing effect; if the finishing effect does not reach the preset threshold, the process parameters of ultrasonic vibration or laser drilling are adjusted, and the finishing process is repeated. After finishing is completed, the key dimensions of the parts are measured online through the machine vision system, and the measurement results are filtered and fused using the Kalman filter algorithm to obtain a high-precision estimate of the part size; the part size estimate is compared with the design size to calculate the size deviation; if the deviation exceeds the preset tolerance range, the injection molding and stamping process parameters and the precision finishing process parameters are adjusted, and the step 1 is returned to be executed again; if the part size deviation and surface quality meet the requirements, the part is transferred to the next process as a qualified product; at the same time, the optimized process parameters are saved in the process database to provide a reference for subsequent batch production.

[0100] Specifically, a part surface defect prediction model is constructed based on the temperature field distribution, flow characteristics and residual stress distribution data obtained from the injection molding process. The support vector machine algorithm is used to identify and locate areas where defects may occur. The Gaussian kernel function is used, the penalty factor C is set to 10, the number of training set samples is 500, and after 10-fold cross validation, the model prediction accuracy reaches 95%; for the identified defective area, a high-definition image of the defective area is obtained through a machine vision system with an image resolution of 1600×1200, and an image segmentation algorithm based on region growth is used to extract the precise contour of the defective area, with a contour extraction accuracy better than 1mm. According to the morphological characteristics of the defect area, defects with an area less than 5mm² and a roundness greater than 9 are removed and repaired by ultrasonic vibration with a frequency of 40kHz and an amplitude of 10μm; defects with an area greater than 5mm² and a roundness less than 9 are processed by laser drilling technology with a wavelength of 532nm, a pulse width of 10ns, and a repetition frequency of 20kHz; during the defect removal and repair process, the surface image of the part is acquired in real time at a speed of 30 frames / second through the machine vision system, and the image quality evaluation algorithm based on the gradient direction histogram is used to judge the finishing effect, and the finishing effect evaluation score threshold is set to 85 points; after the finishing is completed, the key dimensions of the part are measured online with a repeatability of 5μm through the machine vision system, and the Kalman filter algorithm is used to filter and fuse 10 measurement results to obtain the estimated value of the part size, with an estimated accuracy better than 10μm; the estimated value of the part size is compared with the design size, and the size deviation is calculated. The deviation exceeds ±5mm It is judged as out of size, which triggers the adjustment of injection molding and stamping process parameters and precision trimming process parameters, and re-executes defect prediction, defect removal and dimension measurement steps.

[0101] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the concept of the present application. For example, the above features are replaced with the technical features with similar functions disclosed in the present application (but not limited to) to form a technical solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0102] Figure 1 The present invention is a flow chart of an injection molding and stamping molding method for producing 6G antenna hole parts.

[0103] Figure 2 The present invention is a flow chart of step S3 of an injection molding and stamping molding method for producing 6G antenna hole parts. Figure 3The present invention is a flow chart of step S4 of an injection molding and stamping molding method for producing 6G antenna hole parts. DETAILED DESCRIPTION

[0104] In order to make the purpose, technical solution and advantages of the present invention more clear, the present invention is further described in detail below in conjunction with the accompanying drawings and specific implementation methods. The following embodiments can enable technicians in this field to understand the present invention more comprehensively, but the present invention is not limited to the scope of the embodiments.

[0105] like Figure 1-3 , this specific implementation adopts the following technical solution: an injection molding and stamping molding method for producing 6G antenna hole parts, comprising the following steps:

[0106] S1. According to the structural characteristics of the 6G antenna hole parts, computer-aided design software is used to perform three-dimensional modeling of the parts to obtain the geometric parameters and hole distribution information of the parts. Using the obtained geometric model, the injection molding process is numerically simulated through the finite element analysis method to obtain the temperature field and flow field distribution law in the cavity.

[0107] According to the structural characteristics of the 6G antenna hole parts, the parts are 3D modeled to obtain the geometric parameters and hole distribution information of the parts; based on the obtained geometric model and hole distribution information, a finite element analysis model of the injection molding and stamping molding process is constructed; based on the constructed finite element analysis model, the process parameters of the injection molding and stamping molding process are set, including injection temperature, injection pressure, and stamping speed; according to the set process parameters, the injection molding and stamping molding process is numerically simulated to obtain the temperature field and flow field distribution law in the cavity; according to the temperature field distribution law obtained by numerical simulation calculation, it is judged whether the temperature distribution on the surface of the part is uniform during the injection molding and stamping molding process, if not, the injection temperature and cooling time are adjusted; according to the flow field distribution law obtained by numerical simulation calculation, it is judged whether the melt filling is smooth during the injection molding and stamping molding process, if not, the injection pressure and runner design are adjusted; according to the adjusted process parameters, the numerical simulation calculation of the injection molding and stamping molding process is re-performed until the temperature field and flow field distribution that meet the requirements are obtained.

[0108] Specifically, the 6G antenna hole parts were 3D modeled using SolidWorks software. The geometric dimensions of the parts were obtained by measurement. The hole diameter was 8mm, the spacing was 2mm, and the holes were distributed in a rectangular array. The CAD model was imported into ANSYS and tetrahedral meshing was used. The unit size was 1mm and the number of meshes was about 500,000. ABS plastic was used as the material, with a density of 0.5g / cm³ and an elastic modulus of 25GPa.

[0109] According to manufacturing experience, the initial setting of injection temperature is 230℃, injection pressure is 80MPa, and punching speed is 5mm / s. Through numerical simulation, it is found that the surface temperature of the part is unevenly distributed, the temperature in the center area is higher, and the maximum temperature difference is 15℃. Therefore, the injection temperature is adjusted to 220℃, and the cooling time is extended by 2s. At the same time, it is found that the filling near the hole is not smooth, and there is a risk of underfilling. After analysis, it is found that the hole hinders the flow of the melt, so the injection pressure is increased to 100MPa, and the gate and runner design are optimized to make the melt flow around. After 3 iterative calculations, qualified temperature field and flow field results are finally obtained, the temperature difference is controlled within 5℃, and the filling is full and defect-free.

[0110] S2. In order to solve the problem of uneven temperature field distribution shown in the simulation results, multi-point temperature control technology is adopted. Multiple temperature sensors and heaters are arranged at key positions of the mold cavity. The temperature of each area is adjusted in real time through a closed-loop feedback control algorithm to ensure uniform temperature distribution in the cavity and eliminate local overheating or underheating.

[0111] According to the structural characteristics of the mold cavity and the temperature distribution requirements, the key positions are determined, and temperature sensors and heaters are arranged at the key positions. The temperature sensors are used to collect the temperature data of each area in real time, and the heaters are used for local heating. After the temperature sensor collects the real-time temperature data of each area, the temperature data is transmitted to the closed-loop control system. The closed-loop control system determines whether the temperature of each area meets the requirements based on the preset target temperature distribution. If the temperature of a certain area is lower than the target temperature, the closed-loop control system sends a heating instruction to the heater of the area to increase the temperature of the area. If the temperature of a certain area is higher than the target temperature, the closed-loop control system sends a stop heating or cooling instruction to the heater of the area to reduce the temperature of the area.

[0112] The closed-loop control system uses the PID control algorithm to calculate the heating power and heating time of each heater according to the real-time temperature data and target temperature distribution of each area, generates accurate temperature control instructions, and realizes accurate adjustment of the temperature of each area. During the temperature control process, the closed-loop control system also needs to dynamically adjust the parameters of the PID control algorithm according to the real-time temperature data fed back by the temperature sensor to adapt to the changes in the heat conduction characteristics inside the mold cavity and improve the accuracy and response speed of temperature control. According to the temperature status and control effect of each area, the closed-loop control system uses an adaptive control algorithm to optimize the temperature control strategy, automatically adjust the target temperature and control parameters of each area, and improve the efficiency and energy utilization of temperature control while meeting the temperature uniformity requirements. During the temperature control process, the closed-loop control system monitors the temperature change trend of each area in real time, and uses a predictive control algorithm to predict the risk of temperature fluctuation in advance. When it is predicted that a certain area may have abnormal temperature, the temperature of the adjacent area is adjusted in advance to prevent local overheating or underheating, and ensure that the temperature in the mold cavity is always in a uniform and stable state.

[0113] Specifically, according to the structural characteristics and temperature distribution requirements of the mold cavity, PT100 platinum resistance temperature sensors and ceramic heaters are arranged at key positions such as the four corners and center of the cavity; the temperature sensor collects temperature data of each area every 1 second and transmits it to the closed-loop control system based on the STM32 single-chip microcomputer. The control system calculates the PWM duty cycle of each heater based on the preset target temperature distribution (such as 180°C for the four corner areas and 200°C for the center area). For example, according to the temperature deviation and PID parameters P=5, I=2, D=1, when the temperature of the corresponding area is lower than the target temperature, the PWM duty cycle is 80%; when it is higher than the target temperature, the PWM duty cycle is 30%; at the end of each temperature sampling cycle of the control system, the PID parameters are dynamically adjusted using the fuzzy adaptive algorithm according to the average temperature deviation within the cycle, and the temperature change trend of the next cycle is predicted based on the Kalman filter algorithm. The heating power of the adjacent areas is adjusted 5 seconds in advance to control the temperature fluctuation amplitude within ±2°C to ensure the uniformity and stability of the temperature distribution in the mold cavity.

[0114] S3. Based on the optimized temperature field distribution, the gating system and runner are designed and optimized to improve the melt flow balance. Specifically, the temperature field distribution data is used to determine the area with uneven heat distribution in the cavity, and the layout of the gating system and runner is adjusted accordingly. The melt filling process under different casting schemes is analyzed through computer fluid dynamics simulation to obtain the optimal casting position and runner cross-sectional size, so that the flow resistance distribution of the melt in the cavity tends to be uniform, avoiding accumulation or defects around the holes.

[0115] According to the optimized temperature field distribution data, the area with uneven heat distribution in the cavity is judged, and the parts that need to be optimized are determined; computer fluid dynamics simulation technology is used to simulate the melt filling process of different casting systems and runner layout schemes to obtain the flow resistance distribution under each scheme; by comparing the simulation results of different schemes, the best casting position and runner cross-sectional size combination is determined to make the flow resistance distribution of the melt in the cavity most uniform; if the simulation results show that the flow resistance in some areas is still large, the casting system and runner layout of the area are adjusted in a targeted manner, and the simulation verification is re-performed; according to the casting system and runner layout scheme obtained by simulation optimization, the actual casting process parameters are set, including casting temperature and casting speed; during the trial production process, the quality of the finished castings is tested, focusing on whether there are defects near the area with uneven heat distribution in the cavity, and collecting relevant data; according to the trial production data feedback, the casting system and runner layout scheme are further optimized and adjusted until the melt flow balance and casting finished product quality requirements are met.

[0116] Specifically, firstly, according to the optimized temperature field distribution data, the finite element analysis software ANSYS was used to analyze the temperature distribution in the cavity. By comparing the temperature gradients of various areas, it was found that the temperature gradients in the upper left and lower right corners of the cavity were large, and the heat distribution was uneven, which needed to be optimized. Then, the CFD software Fluent was used to simulate the melt filling process of different gating systems and runner layout schemes. By solving the Navier-Stokes equation, the flow velocity and pressure distribution cloud maps under each scheme were obtained, and the flow resistance of each area was calculated. By comparing three different combinations of pouring positions and runner cross-sectional dimensions, it was found that one of the schemes had the most uniform flow resistance distribution, with a standard deviation of only 12 MPa. However, the flow resistance in the lower right corner still reached 35 MPa. Targetedly, the gate diameter of this area was increased from the original 2 cm to 3 cm, and the runner cross-sectional area was increased by 20%. The simulation verification was repeated, and the flow resistance was reduced to 29 MPa. According to the optimization plan, the pouring temperature was set to 720℃ and the pouring speed was set to 5m / s. A trial production of 100 castings was carried out, with a focus on testing the lower right corner area. Two castings were found to have shrinkage defects with a defect rate of 2%. After secondary optimization, the pouring temperature was increased to 730℃, the pouring speed was reduced to 2m / s, and the defect rate was reduced to 5%, meeting the requirements of melt flow balance and finished product quality.

[0117] S4. Aiming at the influence of cavity temperature and flow field distribution on melt viscosity, an intelligent injection molding process parameter optimization method is adopted; according to the rheological properties of the part material, a temperature-viscosity-pressure mathematical model is established; the model adopts a polynomial fitting method, with temperature and pressure as independent variables and viscosity as the dependent variable, and obtains a specific functional relationship through fitting experimental data; the molding quality under different process parameter combinations is predicted and evaluated through machine learning algorithms, and the optimal injection molding temperature, pressure and time parameters are obtained to achieve stable control of the part molding cycle.

[0118] According to the real-time data of the injection molding process, the temperature, pressure and viscosity parameters of the melt are obtained, and a three-dimensional mathematical model of temperature-viscosity-pressure is established; the polynomial fitting method is adopted, with temperature and pressure as independent variables and viscosity as dependent variable, and a large amount of experimental data is fitted to obtain the specific functional relationship between temperature, pressure and viscosity; the fitted mathematical model is used as the input of the machine learning algorithm, and models such as support vector machine SVM, random forest and neural network are trained to predict and evaluate the molding quality under different process parameter combinations; according to the prediction results of the machine learning model, optimization methods such as genetic algorithm are used to search for the optimal injection temperature and pressure. The temperature, pressure and time parameters are combined to obtain an intelligent injection molding process parameter optimization solution; the optimized process parameters are input into the injection molding machine control system, and the temperature, pressure, time and other parameters in the injection molding process are adjusted in real time through closed-loop feedback control to ensure that the melt viscosity is within the optimal range; during the injection molding process, real-time data such as temperature, pressure and viscosity are continuously collected, and the mathematical model and machine learning algorithm are updated and optimized online to continuously improve the accuracy and stability of process parameter optimization; through integration with the injection molding machine control system, the full process of injection molding is intelligently controlled, the part molding cycle is shortened, and product quality and production efficiency are improved.

[0119] Specifically, during the injection molding process, the temperature, pressure, viscosity and other parameters of the melt are collected in real time through sensors, where the temperature range is 180℃~220℃, the pressure range is 50MPa~80MPa, and the viscosity range is 100Pa·s~500Pa·s. Using the polynomial fitting method, with temperature and pressure as independent variables and viscosity as the dependent variable, 500 sets of experimental data are fitted to obtain the functional relationship between temperature, pressure and viscosity: η=05T²-8P²+1TP+200, where η is viscosity, T is temperature, and P is pressure. The mathematical model is input into machine learning models such as support vector machine, random forest and BP neural network, and the molding quality data under 5000 sets of different process parameter combinations are used as training sets to train the model, and the average prediction accuracy can reach more than 95%. On this basis, a genetic algorithm is used to search for the optimal combination of process parameters, with molding cycle and product qualification rate as the optimization targets, crossover probability of 8, mutation probability of 1, population size of 100, and the optimal injection temperature of 205℃, pressure of 65MPa, and holding time of 10s after 50 generations of evolution. The optimized process parameters are input into the injection molding machine control system, and the injection molding process parameters are adjusted in real time through the PID closed-loop control algorithm. The control accuracy can reach ±5℃, ±1MPa, and ±1s, ensuring that the melt viscosity is stable at about 300Pa·s; at the same time, real-time data is continuously collected to update the mathematical model and machine learning algorithm online, and an iterative optimization is performed after every 1,000 products are produced to continuously improve the accuracy and stability of process parameter optimization. Through the above intelligent control method, the injection molding cycle can be shortened by 15%, the product qualification rate can be increased by 5%, and the production efficiency can be increased by 20%.

[0120] S5. In order to reduce the difference between batches, various sensors such as pressure, temperature, displacement, etc. are installed on the injection molding machine, and online quality monitoring and intelligent feedback control technology are introduced; key parameter information of the molding process is collected in real time, and the production data of different batches are mined and compared through big data analysis methods, so as to timely discover and correct process deviations and ensure the consistency of part quality; at the same time, by analyzing the difference data between batches, the key factors affecting the residual stress distribution are identified to provide a basis for subsequent residual stress control.

[0121] According to the requirements of the injection molding process, determine the key parameters of pressure, temperature, and displacement that need to be monitored, select sensors of appropriate models and specifications, and install them at key locations of the injection molding machine; obtain parameter information collected by the sensor in real time through the data acquisition module, and pre-process the data, including denoising and normalization, to obtain standardized parameter data; store the standardized parameter data in the big data platform, and use data mining algorithms such as time series analysis and cluster analysis to compare and analyze production data from different batches to identify the normal range and abnormal deviations of each parameter; if an abnormal deviation is found in a batch of parameters, an early warning is triggered, and possible causes are matched through the expert knowledge base, and corrections are given Suggestions are given to guide operators to adjust process parameters in time; statistical analysis is performed on batch difference data, the correlation between each parameter and the residual stress distribution of parts is calculated, the key factors that have a greater impact on residual stress are identified, and a factor weight list is formed; according to the factor weight list, a residual stress distribution prediction model is established, and the parameter data collected in real time is input to predict the residual stress distribution of molded parts. When the prediction result exceeds the allowable range, an early warning is triggered and optimization suggestions are given; the optimized process parameters are written into the injection molding machine control system, and parameter adaptive optimization is achieved through closed-loop feedback control, which continuously reduces batch differences, improves part quality consistency, and continuously accumulates an optimization knowledge base.

[0122] Specifically, according to the requirements of the injection molding process, Kistler 6152A pressure sensor is selected to monitor the injection pressure, with a range of 0-2000bar and an accuracy of ±5%FS; Pt100 temperature sensor is selected to monitor the barrel temperature, with a range of 0-400℃ and an accuracy of ±5℃; KEYENCE IL-300 laser displacement sensor is selected to monitor the injection displacement, with a range of 0-800mm and an accuracy of ±02%FS. The sensors were installed at key positions such as the barrel and mold of the injection molding machine. The data acquisition module based on LabVIEW was used to collect sensor data in real time at a frequency of 1kHz, and the 5th-order Butterworth low-pass filter was used for denoising and maximum and minimum value normalization. The processed data was stored in the Hadoop big data platform. The ARIMA time series model and K-means clustering algorithm were used to compare and analyze different batches of data, and the normal ranges of pressure, temperature, and displacement were identified to be 800-1200bar, 180-220℃, and 300-500mm, respectively. The abnormal deviation threshold was ±5%. If an abnormality was found, an early warning would be triggered. Rule matching was performed through the expert knowledge base based on Drools, and corrective suggestions such as "excessive injection pressure may be due to gate blockage, and it is recommended to check the gate and clean it" were given. The Pearson correlation coefficient method was used to calculate the correlation between each parameter and the residual stress of the part, and it was found that injection pressure, barrel temperature, and holding time were the key factors affecting residual stress, with weights of 6, 3, and 1, respectively. Based on this, a residual stress prediction model based on BP neural network was established. Real-time parameter data was input to predict the residual stress distribution. When the prediction result exceeded 50MPa, an early warning was triggered. The process parameters were optimized through genetic algorithm, and the optimal parameter combination was obtained as injection pressure 1000bar, barrel temperature 200℃, and holding time 10s. The optimal parameter combination was written into the Siemens S7-1500 PLC control system, and the closed-loop control of the injection molding machine parameters was realized through Profinet industrial Ethernet. The optimization knowledge was continuously accumulated in the MongoDB knowledge base to form an intelligent optimization closed loop of the injection molding process.

[0123] S6. To solve the warping problem caused by uneven residual stress distribution inside parts, a multi-step injection molding process is adopted based on the above-mentioned optimized process parameters and batch difference analysis results. The process includes the following steps: first, pre-filling, then segmented filling according to the preset time interval and pressure gradient, and finally pressure holding and cooling; through this layered filling and sequential cooling method, the flow and solidification process of the melt in the cavity is controlled, so that the residual stress distribution inside the part is more uniform and the deformation risk is reduced.

[0124] According to the structural characteristics and material properties of the parts, a numerical simulation model of the injection molding process is established, and the optimal combination of injection molding process parameters, including injection temperature, injection speed, holding pressure and time, is determined through simulation analysis. The injection molding process is monitored in real time by machine vision technology, and the position of the melt flow front and the cavity filling state are identified through image processing algorithms. When it is judged that the filling is uneven, the injection speed and pressure distribution are dynamically adjusted to achieve optimized control of the melt filling process; the internal temperature distribution data of the parts are obtained, and the residual stress distribution is calculated by the finite element analysis method in combination with the pvT characteristic curve of the material. When the unevenness of the stress distribution exceeds the preset threshold, the injection molding process parameters are adjusted, such as extending the cooling time or optimizing the cooling system layout; a data-driven model of injection molding product quality and process parameters is established, and machine learning algorithms such as support vector machines or neural networks are used to obtain a quality prediction model through training with a large amount of historical data. The injection molding process settings are optimized according to the model prediction results to improve the product qualification rate.

[0125] The pressure, temperature and other parameters of the injection molding process are collected and statistically analyzed in real time, and a multivariate linear regression model is constructed to explore the quantitative relationship between each parameter and the deformation of the part, determine the key control parameters, and guide the optimization of the injection molding process. For different batches of raw materials, non-destructive testing methods such as infrared spectroscopy are used to characterize their performance parameters, such as viscosity, melt flow rate, etc. The test data are input into the quality prediction model to dynamically adjust the injection molding process parameters of the batch to reduce the impact of batch differences on product quality. In the molding process, external fields such as vibration or ultrasound are introduced to induce the melt to produce a uniform micro-orientation by adjusting parameters such as the external field frequency and energy, thereby improving the residual stress distribution inside the part and reducing deformation.

[0126] Specifically, according to the structural characteristics and material properties of the parts, Moldflow software was used to establish a three-dimensional numerical simulation model of the injection molding process. Through orthogonal experimental design, factors such as injection temperature of 280-320℃, injection speed of 60-120mm / s, holding pressure of 60-120MPa and holding time of 5-15s were selected. Each factor had 3 levels, with a total of 27 experiments. The simulation analysis obtained the optimal process parameter combination of injection temperature of 300℃, injection speed of 90mm / s, holding pressure of 90MPa and holding time of 10s. A high-speed camera system based on CCD sensor was used to monitor the injection molding process in real time. The image acquisition frequency was 1000 frames / second. The image segmentation algorithm based on the regional growing method was used to extract the position of the melt flow front and calculate the cavity filling rate. When the filling rate deviation exceeded 5%, the injection speed and pressure distribution were dynamically adjusted through the PID control algorithm to achieve optimal control of the melt filling process.

[0127] Four thermocouple sensors are arranged in the cavity to collect the internal temperature distribution data of the part. Combined with the pvT characteristic curve of the material, the finite element analysis software ANSYS is used to divide 100,000 tetrahedral units and calculate the residual stress distribution cloud map. When the stress distribution unevenness exceeds 20%, the injection molding process parameters are optimized by the response surface method, the cooling time is extended to 30s, and the cooling water channel layout is optimized to reduce the cooling water channel spacing by 20% to improve the residual stress distribution. A data-driven model of the dimensional accuracy and process parameters of injection molding products is established. 500 samples are randomly selected, 20 key dimensions of each sample are measured, and 10,000 sets of dimensional data and corresponding process parameter settings are obtained. The support vector machine algorithm is used to train the dimensional accuracy prediction model, and the prediction accuracy is 95% through cross-validation. According to the model prediction results, the injection molding process parameters are optimized to improve the product qualification rate to more than 98%.

[0128] The pressure, temperature and other parameters of the injection molding process are collected in real time. 100 sets of data are collected in each cycle, and a total of 100,000 sets of data are collected to construct a multivariate linear regression model. Through variance analysis, it is determined that pressure and temperature are the key factors affecting the deformation of parts. For every 10MPa increase in pressure, the deformation increases by 5mm, and for every 10℃ increase in temperature, the deformation increases by 3mm. For different batches of PP raw materials, the melt flow rate is determined by infrared spectroscopy, and the viscosity is determined by rotational rheometer. The performance parameter deviation of different batches of materials is within 5%. The test data is input into the quality prediction model to dynamically adjust the injection molding pressure and temperature. The pressure adjustment range is ±5MPa, and the temperature adjustment range is ±10℃ to reduce the impact of batch differences on product quality. During the molding process, 4 piezoelectric ceramic vibrators are arranged on the inner wall of the cavity with a vibration frequency of 20kHz and a vibration energy of 100W. Through finite element analysis, the layout of the vibrators is optimized to make the melt evenly stressed and induce the melt to produce uniform micro-orientation. The unevenness of residual stress distribution is reduced to less than 10%, effectively reducing part deformation.

[0129] S7. To further improve the dimensional accuracy and surface quality of parts, a precision finishing process is introduced after injection molding and stamping. Based on the temperature field distribution, flow characteristics and residual stress distribution data obtained in the previous steps, the areas where defects may occur on the surface of the parts are determined. For these areas, ultrasonic vibration or laser drilling and other technologies are used to remove and repair tiny defects on the surface of the parts. The key dimensions of the parts are measured online and feedback controlled through a machine vision inspection system, and the finishing process parameters are continuously optimized to ultimately meet the high-precision requirements of 6G antenna components.

[0130] Based on the temperature field distribution, flow characteristics and residual stress distribution data obtained from the injection molding process, a part surface defect prediction model is constructed, and the support vector machine algorithm is used to identify and locate areas where defects may occur; for the identified defective areas, a high-definition image of the defective area is obtained through a machine vision system, and an image segmentation algorithm is used to extract the precise contour of the defective area; based on the morphological characteristics of the defective area, it is determined to use ultrasonic vibration or laser drilling technology for defect removal and repair; if the defective area is small and regular in shape, ultrasonic vibration technology is used; if the defective area is large and irregular in shape, laser drilling technology is used; in the process of defect removal and repair, the part surface image is obtained in real time through a machine vision system, and an image quality evaluation algorithm is used to judge the finishing effect; if the finishing effect does not reach the preset threshold, the process parameters of ultrasonic vibration or laser drilling are adjusted, and the finishing process is repeated. After finishing is completed, the key dimensions of the parts are measured online through the machine vision system, and the measurement results are filtered and fused using the Kalman filter algorithm to obtain a high-precision estimate of the part size; the part size estimate is compared with the design size to calculate the size deviation; if the deviation exceeds the preset tolerance range, the injection molding and stamping process parameters and the precision finishing process parameters are adjusted, and the step 1 is returned to be executed again; if the part size deviation and surface quality meet the requirements, the part is transferred to the next process as a qualified product; at the same time, the optimized process parameters are saved in the process database to provide a reference for subsequent batch production.

[0131] Specifically, a part surface defect prediction model is constructed based on the temperature field distribution, flow characteristics and residual stress distribution data obtained from the injection molding process. The support vector machine algorithm is used to identify and locate areas where defects may occur. The Gaussian kernel function is used, the penalty factor C is set to 10, the number of training set samples is 500, and after 10-fold cross validation, the model prediction accuracy reaches 95%; for the identified defective area, a high-definition image of the defective area is obtained through a machine vision system with an image resolution of 1600×1200, and an image segmentation algorithm based on region growth is used to extract the precise contour of the defective area, with a contour extraction accuracy better than 1mm. According to the morphological characteristics of the defect area, defects with an area less than 5mm² and a roundness greater than 9 are removed and repaired by ultrasonic vibration with a frequency of 40kHz and an amplitude of 10μm; defects with an area greater than 5mm² and a roundness less than 9 are processed by laser drilling technology with a wavelength of 532nm, a pulse width of 10ns, and a repetition frequency of 20kHz; during the defect removal and repair process, the surface image of the part is acquired in real time at a speed of 30 frames / second through the machine vision system, and the image quality evaluation algorithm based on the gradient direction histogram is used to judge the finishing effect, and the finishing effect evaluation score threshold is set to 85 points; after the finishing is completed, the key dimensions of the part are measured online with a repeatability of 5μm through the machine vision system, and the Kalman filter algorithm is used to filter and fuse 10 measurement results to obtain the estimated value of the part size, with an estimated accuracy better than 10μm; the estimated value of the part size is compared with the design size, and the size deviation is calculated. The deviation exceeds ±5mm It is determined that the size is out of tolerance, which triggers the adjustment of the injection molding and stamping process parameters and the precision trimming process parameters, and re-executes the steps of defect prediction, defect removal and size measurement. The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in this application is not limited to the technical solution formed by a specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalent features without departing from the concept of the present application. For example, the above-mentioned features are replaced with the technical features with similar functions disclosed in this application (but not limited to) to form a technical solution.

Claims

1. An injection molding method for producing 6G antenna hole parts, characterized in that: The following steps are involved: S1. According to the structural characteristics of the 6G antenna with holes, the computer-aided design software is used to perform three-dimensional modeling of the parts, and the geometric parameters and hole distribution information of the parts are obtained. The obtained geometric model is used to perform numerical simulation of the injection molding process through the finite element analysis method to obtain the temperature field and flow field distribution law in the cavity; S2. In view of the uneven temperature field distribution problem shown in the simulation results, multi-point temperature control technology is used to arrange multiple temperature sensors and heaters at key positions of the mold cavity. The temperature of each area is adjusted in real time through a closed-loop feedback control algorithm to ensure uniform temperature distribution in the cavity and eliminate local overheating or underheating; S3. Based on the optimized The temperature field distribution is used to optimize the gating system and runner design to improve the melt flow balance. The temperature field distribution data is used to determine the area with uneven heat distribution in the cavity, and the layout of the gating system and runner is adjusted accordingly. The melt filling process under different gating schemes is analyzed through computer fluid dynamics simulation to obtain the optimal gating position and runner cross-sectional size, so that the flow resistance distribution of the melt in the cavity tends to be uniform, avoiding accumulation or defects around the holes; S4. In view of the influence of the temperature and flow field distribution in the cavity on the melt viscosity, an intelligent injection molding process parameter optimization method is adopted, and a temperature-viscosity-pressure mathematical model is established according to the rheological properties of the part material. The model adopts a polynomial fitting method, with temperature and pressure as independent variables and viscosity as the dependent variable. The specific functional relationship is obtained by fitting experimental data. The molding quality under different process parameter combinations is predicted and evaluated by machine learning algorithms to obtain the optimal injection molding temperature, pressure and time parameters, thereby achieving stable control of the part molding cycle. S5. In order to reduce the difference between batches, multiple sensors such as pressure, temperature and displacement are installed on the injection molding machine, and online quality monitoring and intelligent feedback control technology are introduced to collect key parameter information of the molding process in real time. The production data of different batches are mined and compared through big data analysis methods to timely discover and correct process deviations and ensure zero The consistency of part quality is ensured. At the same time, by analyzing the batch difference data, the key factors affecting the residual stress distribution are identified to provide a basis for the subsequent residual stress control; S6. In view of the warping deformation problem caused by the uneven distribution of residual stress inside the parts, based on the aforementioned optimized process parameters and batch difference analysis results, a multi-step injection molding process is adopted. The process includes the following steps: first, pre-filling is performed, then segmented filling is performed according to the preset time interval and pressure gradient, and finally, pressure holding and cooling are performed. Through this layered filling and sequential cooling method, the flow and solidification process of the melt in the cavity is controlled, so that the residual stress distribution inside the part is more uniform and the deformation risk is reduced;S7. To further improve the dimensional accuracy and surface quality of parts, a precision finishing process is introduced after injection molding and stamping. Based on the temperature field distribution, flow characteristics and residual stress distribution data obtained in the previous steps, the areas where defects may occur on the surface of the parts are determined. For these areas, ultrasonic vibration or laser drilling technology is used to remove and repair the tiny defects on the surface of the parts. The key dimensions of the parts are measured online and feedback controlled through the machine vision inspection system, and the finishing process parameters are continuously optimized to finally meet the high-precision requirements of 6G antenna components. ; 2. The method for producing a 6G antenna with hole parts by injection molding according to claim 1, characterized in that: The step S1 specifically includes: According to the structural characteristics of the 6G antenna hole parts, the parts are 3D modeled to obtain the geometric parameters and hole distribution information of the parts; Based on the obtained geometric model and hole distribution information, a finite element analysis model of the injection molding process is constructed; According to the constructed finite element analysis model, the process parameters of the injection molding process are set, including injection temperature, injection pressure and stamping speed; According to the set process parameters, the numerical simulation calculation of the injection molding process is carried out to obtain the temperature field and flow field distribution law in the cavity; According to the temperature field distribution law obtained by numerical simulation, it is judged whether the temperature distribution on the surface of the part is uniform during the injection molding process. If it is not uniform, the injection temperature and cooling time are adjusted; According to the flow field distribution law obtained by numerical simulation, it is judged whether the melt filling is smooth during the injection molding process. If not, the injection pressure and runner design are adjusted; According to the adjusted process parameters, the numerical simulation calculation of the injection molding process is carried out again until the temperature field and flow field distribution that meet the requirements are obtained, and the injection molding process optimization of the 6G antenna hole parts is completed.

3. The method for producing a 6G antenna with hole parts by injection molding according to claim 1, characterized in that: The step S2 specifically includes: According to the structural characteristics of the mold cavity and the temperature distribution requirements, the key positions are determined and temperature sensors and heaters are arranged at the key positions. The temperature sensors are used to collect temperature data of each area in real time, and the heaters are used for local heating. After the temperature sensor collects the real-time temperature data of each area, it transmits the temperature data to the closed-loop control system. The closed-loop control system determines whether the temperature of each area meets the requirements based on the preset target temperature distribution; If the temperature of a certain area is lower than the target temperature, the closed-loop control system sends a heating instruction to the heater in that area to increase the temperature of that area; If the temperature of a certain area is higher than the target temperature, the closed-loop control system sends a stop heating or cooling instruction to the heater in that area to reduce the temperature of that area; The closed-loop control system uses the PID control algorithm to calculate the heating power and heating time of each heater based on the real-time temperature data and target temperature distribution of each area, generates accurate temperature control instructions, and realizes accurate adjustment of the temperature of each area; During the temperature control process, the closed-loop control system also needs to dynamically adjust the parameters of the PID control algorithm based on the real-time temperature data fed back by the temperature sensor to adapt to the changes in the heat conduction characteristics inside the mold cavity and improve the accuracy and response speed of temperature control; The closed-loop control system uses an adaptive control algorithm to optimize the temperature control strategy according to the temperature status and control effect of each area, automatically adjusts the target temperature and control parameters of each area, and improves the efficiency of temperature control and energy utilization while meeting the temperature uniformity requirements; During the temperature control process, the closed-loop control system monitors the temperature change trend of each area in real time, and uses a predictive control algorithm to predict the risk of temperature fluctuations in advance. When it is predicted that a certain area may have abnormal temperature, the temperature of the adjacent area is adjusted in advance to prevent local overheating or underheating, ensuring that the temperature in the mold cavity is always in a uniform and stable state.

4. The injection molding and stamping molding method for producing 6G antenna hole parts according to claim 1, characterized in that: The step S3 specifically includes: According to the optimized temperature field distribution data, determine the area with uneven heat distribution in the cavity and determine the area that needs to be optimized; Computer fluid dynamics simulation technology is used to simulate the melt filling process of different gating systems and runner layout schemes to obtain the flow resistance distribution under each scheme; By comparing the simulation results of different schemes, the best combination of pouring position and runner cross-section size is determined to make the flow resistance of the melt in the cavity most evenly distributed; If the simulation results show that the flow resistance in certain areas is still large, the gating system and runner layout of the area should be adjusted accordingly, and the simulation verification should be repeated; According to the gating system and runner layout scheme obtained through simulation optimization, the actual gating process parameters are set, including the gating temperature and gating speed; During the trial production process, the quality of the finished castings is tested, with a focus on whether there are defects near the area of ​​uneven heat distribution in the cavity, and relevant data is collected; Based on the feedback from the trial production data, the gating system and runner layout are further optimized and adjusted until the melt flow balance and casting product quality requirements are met.

5. The method for producing a 6G antenna with hole parts by injection molding according to claim 1, characterized in that: The step S4 specifically includes: According to the real-time data of the injection molding process, the temperature, pressure and viscosity parameters of the melt are obtained, and a three-dimensional mathematical model of temperature-viscosity-pressure is established; The polynomial fitting method is adopted, with temperature and pressure as independent variables and viscosity as dependent variable, and a large amount of experimental data is fitted to obtain the specific functional relationship between temperature, pressure and viscosity; The fitted mathematical model is used as the input of the machine learning algorithm to train the support vector machine, random forest and neural network models to predict and evaluate the molding quality under different process parameter combinations; According to the prediction results of the machine learning model, the genetic algorithm optimization method is used to search for the optimal combination of injection temperature, pressure and time parameters to obtain an intelligent injection molding process parameter optimization solution; The optimized process parameters are input into the injection molding machine control system. Through closed-loop feedback control, the temperature, pressure and time parameters of the injection molding process are adjusted in real time to ensure that the melt viscosity is within the optimal range. During the injection molding process, real-time data on temperature, pressure and viscosity are continuously collected to update and optimize mathematical models and machine learning algorithms online, continuously improving the accuracy and stability of process parameter optimization; By integrating with the injection molding machine control system, the whole process of injection molding can be intelligently controlled, the part molding cycle can be shortened, and product quality and production efficiency can be improved.

6. The method for producing a 6G antenna hole component by injection molding according to claim 1, characterized in that: The step S5 specifically includes: According to the requirements of the injection molding process, determine the key parameters of pressure, temperature and displacement that need to be monitored, select sensors of appropriate models and specifications, and install them at key locations of the injection molding machine; The parameter information collected by the sensor is acquired in real time through the data acquisition module, and the data is preprocessed, including denoising and normalization, to obtain standardized parameter data; The standardized parameter data is stored in the big data platform, and the production data of different batches are compared and analyzed using time series analysis and cluster analysis data mining algorithms to identify the normal range and abnormal deviation of each parameter; If an abnormal deviation is found in a batch of parameters, an early warning is triggered, and the expert knowledge base is used to match the possible causes and give corrective suggestions to guide operators to adjust process parameters in a timely manner; Conduct statistical analysis on the difference data between batches, calculate the correlation between each parameter and the residual stress distribution of parts, identify the key factors affecting residual stress, and form a factor weight list; According to the factor weight list, a residual stress distribution prediction model is established, and the parameter data collected in real time is input to predict the residual stress distribution of the molded parts. When the prediction result exceeds the allowable range, an early warning is triggered and optimization suggestions are given; The optimized process parameters are written into the injection molding machine control system, and parameter adaptive optimization is achieved through closed-loop feedback control, which continuously reduces batch differences, improves part quality consistency, and continuously accumulates an optimization knowledge base.

7. The method for producing a 6G antenna with hole parts by injection molding according to claim 1, characterized in that: The step S6 specifically includes: According to the structural characteristics and material properties of the parts, a numerical simulation model of the injection molding process is established, and the optimal combination of injection molding process parameters, including injection temperature, injection speed, holding pressure and time, is determined through simulation analysis; The injection molding process is monitored in real time using machine vision technology. The position of the melt flow front and the cavity filling status are identified through image processing algorithms. When uneven filling is detected, the injection speed and pressure distribution are dynamically adjusted to achieve optimal control of the melt filling process. Obtain the internal temperature distribution data of the part, combine it with the material's pvT characteristic curve, calculate the residual stress distribution through finite element analysis, and adjust the injection molding process parameters when the stress distribution unevenness exceeds the preset threshold; Establish a data-driven model for injection molding product quality and process parameters, use machine learning algorithms such as support vector machines or neural networks, obtain a quality prediction model through training with a large amount of historical data, optimize the injection molding process settings based on the model prediction results, and improve the product qualification rate; Real-time collection and statistical analysis of the pressure and temperature parameters of the injection molding process are carried out to build a multivariate linear regression model, explore the quantitative relationship between each parameter and the deformation of the part, determine the key control parameters, and guide the optimization of the injection molding process; For different batches of raw materials, infrared spectroscopy nondestructive testing method is used to characterize their performance parameters, and the test data is input into the quality prediction model to dynamically adjust the injection molding process parameters of the batch to reduce the impact of batch differences on product quality; In the molding process, vibration or ultrasonic external field is introduced to adjust the frequency and energy parameters of the external field to induce the melt to produce a uniform micro-orientation, improve the residual stress distribution inside the part, and achieve the purpose of reducing deformation.

8. The method for producing a 6G antenna hole component by injection molding according to claim 1, characterized in that: The step S7 specifically includes: Based on the temperature field distribution, flow characteristics and residual stress distribution data obtained from the injection molding process, a part surface defect prediction model is constructed, and the support vector machine algorithm is used to identify and locate areas where defects may occur; For the identified defective area, a high-definition image of the defective area is obtained through a machine vision system, and an image segmentation algorithm is used to extract the precise contour of the defective area; According to the morphological characteristics of the defect area, determine whether to use ultrasonic vibration or laser drilling technology to remove and repair the defect; If the defect area is small and regular in shape, ultrasonic vibration technology is used; If the defect area is large and irregular in shape, laser drilling technology is used; During defect removal and repair, the machine vision system is used to obtain real-time images of the part surface, and the image quality assessment algorithm is used to determine the repair effect; If the trimming effect does not reach the preset threshold, the process parameters of ultrasonic vibration or laser drilling are adjusted and the trimming process is repeated; After trimming is completed, the key dimensions of the parts are measured online through a machine vision system, and the Kalman filter algorithm is used to filter and fuse the measurement results to obtain a high-precision estimate of the part dimensions; Compare the estimated part size with the designed size and calculate the dimensional deviation; If the deviation exceeds the preset tolerance range, the injection molding and stamping process parameters and precision finishing process parameters are adjusted, and the surface defect identification is returned and re-executed; If the part's dimensional deviation and surface quality meet the requirements, the part will be transferred to the next process as a qualified product; At the same time, the optimized process parameters are saved in the process database to provide reference for subsequent batch production.

Images