A gradient injection molding method for a Bluetooth earphone case shell

Abstract

The application relates to the field of information technology, in particular to a gradient injection molding method of a Bluetooth earphone case shell, which belongs to the field of injection molding processes of earphone case shells and comprises the following steps: according to a relationship model, initial parameters of material distribution in a gradient injection process are determined, including setting values of temperature, pressure and injection speed; through a finite element analysis tool, the flowing and solidification process of the material in a mold is simulated, and heat conduction performance and mechanical strength of the gradient distribution are predicted; in the injection molding process, the temperature and pressure changes in the mold are monitored in real time through a sensor, and dynamic data of the material flowing and solidifying are acquired; whether the gradient distribution is consistent with the preset model is judged according to the dynamic data; a machine learning algorithm is adopted, a performance prediction model is trained based on historical injection data, the initial setting of the material distribution parameters is optimized, and the production efficiency is improved; through automatic detection equipment, the heat dissipation performance and structural strength of the formed shell are tested, and the consistency and yield of the product are ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of information technology, specifically to the field of injection molding process of earphone case shell, and particularly relates to a gradient injection molding method for Bluetooth earphone case shell. BACKGROUND

[0002] In the gradient injection molding process of Bluetooth earphone case shell, the traditional method uses single material injection, which makes it difficult to meet the requirements of both heat dissipation performance and structural strength. Single material cannot balance the heat conductivity and mechanical strength, poor heat dissipation performance will cause the temperature of internal electronic components to be too high, affecting the service life of the equipment, and insufficient structural strength will reduce the impact resistance of the shell and increase the risk of damage.

[0003] Although the existing gradient injection technology can realize the layered optimization of material performance, it mainly relies on multi-layer mold switching. This process not only increases the complexity of mold design and manufacturing, but also leads to longer production cycle and higher cost, and it is difficult to realize continuous gradient transition of material performance. In addition, the multi-layer mold switching process is prone to poor interface bonding, affecting the mechanical properties and appearance quality of the final product.

[0004] Therefore, there is an urgent need for a new injection molding method that can realize continuous gradient change of material performance in a single mold, to solve the technical contradiction between heat dissipation and structural strength. At the same time, this method also needs to reduce production cost, improve production efficiency, and ensure product consistency and yield on the premise of performance improvement. SUMMARY

[0005] The present application provides a gradient injection molding method for Bluetooth earphone case shell, comprising the following steps:

[0006] S101, using a preset gradient material database, obtaining the thermal conductivity coefficient and mechanical strength data of the material, and establishing a relationship model between material performance and gradient distribution;

[0007] S102, according to the relationship model, determining the initial parameters of material distribution in the gradient injection process, including the set values of temperature, pressure and injection speed;

[0008] S103, through a finite element analysis tool, simulating the flow and solidification process of the material in the mold, and predicting the thermal conductivity performance and mechanical strength of the gradient distribution;

[0009] S104, if the simulation result does not reach the preset performance threshold, adjust the material distribution parameters and perform finite element analysis again until the requirements of heat dissipation and structural strength are met;

[0010] S105, using an intelligent control system to adjust the injection molding parameters in real time within a single mold, realizing continuous gradient transition of material properties, and avoiding poor interface bonding problems;

[0011] S106, during the injection molding process, real-time monitoring of temperature and pressure changes in the mold through sensors to obtain dynamic data of material flow and solidification;

[0012] S107, according to the dynamic data, determine whether the gradient distribution is consistent with the preset model, if not, feedback to the intelligent control system, and real-time optimize the injection molding parameters;

[0013] S108, using machine learning algorithm, based on historical injection data to train performance prediction model, optimize the initial setting of material distribution parameters, and improve production efficiency;

[0014] S109, through automatic detection equipment, test the heat dissipation performance and structural strength of the formed shell, and ensure the consistency and yield of the product.

[0015] The technical scheme provided by the embodiment of the present application can include the following beneficial effects:

[0016] The application discloses a Bluetooth earphone case shell gradient injection molding method for manufacturing a shell product with continuous gradient performance. The method first establishes a relationship model between material performance and gradient distribution, and determines the initial injection molding parameters. Through finite element analysis to simulate the material flow and solidification process, the performance of the gradient distribution is predicted. If the preset threshold is not reached, the parameters are adjusted and reanalyzed. During the injection molding process, the application uses an intelligent control system to adjust the parameters in real time, realizes the continuous gradient transition of the material performance. At the same time, through the sensor to monitor the dynamic data in the mold, and compare with the preset model, real-time optimize the injection molding parameters. In addition, the application also uses machine learning algorithm to optimize the initial parameter setting, and through automatic detection to ensure the product quality. This method can effectively solve the problem of poor interface bonding in traditional injection molding, improve the heat dissipation performance and structural strength of the product, and improve the production efficiency and yield. BRIEF DESCRIPTION OF DRAWINGS

[0017] Fig. 1 The flow chart of the Bluetooth earphone case shell gradient injection molding method of the application.

[0018] Fig. 2 The schematic diagram of the Bluetooth earphone case shell gradient injection molding method of the application.

[0019] Fig. 3 Another schematic diagram of the Bluetooth earphone case shell gradient injection molding method of the application. DETAILED DESCRIPTION

[0020] For further understanding of the present application, the application will be described in detail with reference to the drawings and examples. The following further describes the application with reference to the drawings and examples. It can be understood that the specific examples described herein are only used to explain the related application, and not to limit the application. In addition, it should be noted that, for ease of description, only the parts related to the application are shown in the drawings.

[0021] As Figs. 1-3 The gradient injection molding method of the Bluetooth earphone case shell in the embodiment can specifically include the following steps.

[0022] In step S101, a preset gradient material database is used to obtain the thermal conductivity coefficient and mechanical strength data of the material, and a relationship model of the material performance and the gradient distribution is established.

[0023] The preset gradient material database is obtained, and the gradient material database includes the thermal conductivity coefficient and mechanical strength data of the material. The gradient distribution parameters of the material are obtained, and the gradient distribution characteristics of the material are calculated according to the gradient distribution parameters. A first machine learning algorithm is used to establish a first mapping relationship between the thermal conductivity coefficient and the gradient distribution characteristics. A second machine learning algorithm is used to establish a second mapping relationship between the mechanical strength and the gradient distribution characteristics. If the thermal conductivity coefficient and the mechanical strength have a correlation, the correlation data of the thermal conductivity coefficient and the mechanical strength is obtained, and a comprehensive performance model of the material is generated according to the correlation data. The relationship between the material performance and the gradient distribution characteristics is determined according to the comprehensive performance model. The target gradient distribution parameters are obtained, and the target gradient distribution characteristics are calculated according to the target gradient distribution parameters. The performance parameters of the material under the target gradient distribution characteristics are determined according to the relationship between the material performance and the gradient distribution characteristics.

[0024] Specifically, the thermal conductivity coefficients in the gradient material database usually include typical materials such as base metals and functional ceramics. For example, titanium alloy base may have a low thermal conductivity coefficient, while titanium carbide ceramic has a higher thermal conductivity coefficient. The mechanical strength database records the strength changes of each component at different temperatures, such as titanium alloy having higher tensile strength and fatigue strength at room temperature. The gradient distribution characteristics can be characterized by parameters such as volume fraction and component thickness. For example, in the thermal barrier coating of turbine blades, the thickness of the zirconia ceramic coating gradually decreases from the blade surface to the base, and the volume fraction of the titanium alloy base increases accordingly. This trend can be described by a piecewise function or a continuous function. Through machine learning algorithms such as support vector machines, a mapping relationship between thermal conductivity coefficients and gradient distribution can be established. Taking turbine blades as an example, when the thickness of the ceramic coating increases, the overall thermal conductivity coefficient decreases, thereby improving the thermal insulation effect. At the same time, the neural network model is used to analyze the relationship between mechanical strength and gradient distribution, and it is found that with the increase of the thickness of the ceramic coating, the interface stress concentration phenomenon intensifies, affecting the bonding strength of the coating. In the application of turbine blades, there is indeed a correlation between thermal conductivity and mechanical strength. Although thicker ceramic coatings have better thermal insulation performance, excessive thickness can cause interface peeling. Therefore, a comprehensive performance model needs to be established to ensure sufficient thermal insulation effect while ensuring that the coating has good bonding strength. Through this model, the comprehensive performance under different gradient distributions can be predicted. Another typical application is armor protection materials, in which the gradient change of ceramic and metal matrix affects the propagation of impact energy and determines the overall structural strength. The surface layer with high ceramic content provides good bulletproof performance, while the metal matrix ensures sufficient toughness. Through gradient distribution optimization, the best balance between protection performance and structural reliability can be achieved. According to the analysis results of the comprehensive performance model, the optimal gradient distribution of the material in various application scenarios can be determined. For example, in high temperature working conditions, it may be necessary to increase the thickness of the surface ceramic to provide better thermal protection, while in the case of large mechanical load, it is necessary to appropriately reduce the thickness of the ceramic to improve the structural reliability. This optimization design provides important guidance for the practical application of materials.

[0025] In step S102, according to the relationship model, the initial parameters of the material distribution in the gradient injection molding process are determined, including the set values of temperature, pressure and injection speed.

[0026] The historical data of material distribution in the gradient injection molding process is acquired, and a correlation model of the temperature, the pressure and the injection speed is established; according to the correlation model, the optimal temperature value, pressure value and speed value range of the material distribution are calculated; according to the optimal temperature value, pressure value and speed value range, the initial setting values of the temperature, the pressure and the injection speed are determined; the material distribution state information is acquired, and it is judged whether the material distribution state is abnormal; if the material distribution state is abnormal, the setting values of the temperature, the pressure and the injection speed are adjusted, and the material distribution state is recalculated; the material distribution state is monitored in real time, and it is judged whether the setting values of the temperature, the pressure and the injection speed reach the predetermined target; if the setting values of the temperature, the pressure and the injection speed do not reach the predetermined target, the setting values of the temperature, the pressure and the injection speed are automatically adjusted, and the material distribution state is optimized; the machine learning algorithm is used to learn the adjusted setting values of the temperature, the pressure and the injection speed, and the correlation of the temperature, the pressure and the injection speed in the correlation model is updated.

[0027] Specifically, the historical data of material distribution in the gradient injection molding process includes three core parameters: temperature, pressure, and injection speed. The historical data shows that when the mold cavity temperature is between 180°C and 220°C, the injection pressure is controlled between 70 and 90 MPa, and the injection speed is maintained between 50 and 70 mm / s, the material distribution is the most uniform. By analyzing these data, a correlation model between parameters can be established. When calculating the optimal parameter range in the preset algorithm, a neural network model can be used. Taking polycarbonate as an example, by analyzing a large amount of historical data, it is found that when the mold temperature is 180°C, the injection pressure is 100 MPa, and the injection speed is 60 mm / s, the best material distribution state can be achieved. These parameters serve as initial setting values, providing a basis for subsequent optimization. In actual production, if uneven material distribution is found, such as accumulation at the mold corner, the parameters need to be adjusted in time. For example, increase the injection pressure to 110 MPa and reduce the injection speed to 50 mm / s, which can improve the flow performance of the material at the corner. The real-time monitoring system continuously collects production data through pressure sensors and temperature sensors. When it detects that the temperature in a certain area of the mold is too low, the system will automatically increase the heating power of that area. Machine learning algorithms play an important role in parameter optimization. Taking nylon material as an example, through learning the parameters adjusted multiple times, the system finds that when the mold temperature increases, the injection pressure needs to be correspondingly reduced to maintain good material distribution. This correlation is updated to the model to guide subsequent production. The evaluation of material distribution state uses multiple indicators. For example, the pressure difference in different areas of the mold does not exceed 5%, the temperature fluctuation is controlled within ±3°C, and the wall thickness error of the finished product is less than 2%. These indicators together determine the direction and amplitude of parameter adjustment. When it is found that the wall thickness in a certain place is too thin, the system will appropriately increase the holding time and pressure in that area while ensuring that the material distribution in other areas is not affected. Through continuous optimization and learning, the relationship model becomes more accurate. For example, for polyethylene material, the system can automatically recommend the most suitable process parameter combination according to the structural characteristics of different products, and make fine adjustments in real time during production to ensure the stability of product quality. This intelligent parameter adjustment method greatly improves the efficiency and product quality of gradient injection molding.

[0028] Step S103, simulate the flow and solidification process of the material in the mold through the finite element analysis tool, and predict the thermal conductivity and mechanical strength of the gradient distribution.

[0029] A mathematical model of material flow and solidification is established using a finite element tool to obtain flow state and solidification state data of the material in the mold. According to the flow state and solidification state data, the thermal conductivity performance value and the mechanical strength value inside the material are calculated through a gradient distribution algorithm. For the thermal conductivity performance value and the mechanical strength value, a regression model in a machine learning algorithm is used to predict a gradient distribution map of the material in the mold. If the predicted gradient distribution map matches a preset threshold range, it is determined that the performance value of the material meets the design requirements. If it does not match, the simulation parameters of the flow state and the solidification state are adjusted, and the thermal conductivity performance value and the mechanical strength value are recalculated. According to the adjusted parameters, the flow and solidification process of the material in the mold is simulated again through the finite element tool to obtain new flow state and solidification state data. The new flow state and solidification state data are used to regenerate the gradient distribution map to determine whether the performance value of the material meets the design requirements.

[0030] Specifically, finite element analysis tools play a crucial role in material flow and solidification process analysis. For example, in the production of polycarbonate injection molded parts, by setting the initial material temperature at 230 degrees Celsius and the mold temperature at 80 degrees Celsius, the temperature distribution, velocity field, and pressure field changes during the material flow process can be simulated. During the solidification process, the material transitions from a flow state to a solidification state, which causes volume shrinkage. By analyzing the flow state and solidification state data, the gradient distribution algorithm can calculate the material's internal properties. Taking a polyamide injection molded part as an example, by analyzing the crystallinity and molecular orientation at different locations, the thermal conductivity performance values can be obtained, such as the thermal conductivity coefficient being 0.2 watts per meter Kelvin at the edge of the product and 0.3 watts per meter Kelvin at the center. The mechanical strength values are represented by the tensile strength of 50 megapascals at the edge and 60 megapascals at the center. Machine learning regression models can predict the material gradient distribution. In the production of polypropylene injection molded parts, a support vector regression model is used, with process parameters such as mold temperature and pressure in each region as input variables, and local thermal conductivity coefficient and tensile strength as output variables, to establish a prediction model. Through a large amount of historical data training, the model can predict the performance distribution under different process parameters. The prediction results need to match the pre-set threshold range. For example, the thermal conductivity performance requires a range of 0.2 to 0.4 watts per meter Kelvin, and the mechanical strength requires a range of 40 to 70 megapascals. If the performance value of a certain region exceeds the range, the simulation parameters need to be adjusted. The performance distribution can be improved by increasing the mold temperature in that region or reducing the injection speed. After adjustment, new flow state and solidification state data are obtained through finite element analysis. Through iterative optimization, the final material performance distribution meets the requirements. This method can effectively improve the quality stability of injection molded products, reduce the number of trial molds, and reduce production costs. In practical applications, different materials and products have different performance distribution requirements, and reasonable threshold ranges and optimization strategies need to be set according to specific circumstances. Through this systematic analysis method, the performance distribution of injection molded products can be accurately predicted and controlled, ensuring product quality. At the same time, the accumulated data and experience can be used for the development of similar products in the future, improving development efficiency. This method is particularly suitable for the production of precision injection molded parts with high performance requirements, such as medical devices and optical components.

[0031] Step S104, if the simulation result does not meet the pre-set performance threshold, adjust the material distribution parameters and perform finite element analysis again until the heat dissipation and structural strength requirements are met.

[0032] The result value of the finite element analysis is obtained; the result value is compared with a preset value to obtain a comparison result; if the comparison result indicates that the result value does not reach the preset value, an adjustment amount is calculated, and the material distribution parameter is adjusted according to the adjustment amount to obtain an adjusted material distribution parameter; the finite element analysis is re-performed according to the adjusted material distribution parameter to obtain a new result value; it is judged whether the new result value reaches the preset value; if the new result value reaches the preset value, the material distribution parameter meeting the heat dissipation and structural strength is output; if the new result value does not reach the preset value, the steps of adjusting the material distribution parameter, re-performing the finite element analysis, and judging whether the new result value reaches the preset value are repeatedly executed until the new result value reaches the preset value.

[0033] Specifically, finite element analysis is an effective numerical method for studying complex engineering problems and is widely used in material design. For example, in the design of a heat sink, the temperature distribution, heat flux density, and other result values can be obtained through finite element analysis. If the maximum temperature on the surface of the heat sink is eighty-five degrees and the preset safety threshold is seventy degrees, it indicates that the current design scheme does not meet the heat dissipation requirements. At this time, the adjustment amount needs to be calculated, such as increasing the heat dissipation area or changing the fin spacing. The determination of the adjustment amount is based on the relationship between the temperature difference and the thermal resistance. The larger the temperature difference, the greater the adjustment required. The adjustment of the material distribution parameter involves multiple aspects. For example, in the case of a composite structural part, if the stress analysis shows that the stress concentration in a certain area exceeds the preset value, the stress distribution can be optimized by adjusting parameters such as fiber content and arrangement direction. For example, the fiber content is adjusted from thirty percent to forty percent, or the fiber direction is adjusted from zero degrees to plus or minus forty-five degrees. After each adjustment, the finite element analysis is re-performed to evaluate the performance of the new scheme. The judgment of the result value needs to consider multiple indicators. For example, in the case of an electronic product shell, both structural strength and heat dissipation performance need to be met. Assuming that the shell material is modified engineering plastic, the maximum deformation obtained through finite element analysis is zero point eight millimeters, and the thermal deformation temperature is one hundred and twenty degrees. If the preset values are one millimeter and one hundred degrees, respectively, it indicates that the strength indicator meets the requirements but the heat dissipation performance needs to be improved. The performance can be optimized by adding thermal conductive fillers or adjusting the wall thickness distribution. The optimization process is repeated until the design goal is achieved. For example, in the case of an automobile instrument panel support, the first analysis shows that the rigidity is insufficient, and the optimization is performed by increasing the number of rib plates and adjusting the rib plate layout. After each iteration analysis, new stress distribution and deformation data are obtained, which are compared with the preset rigidity and strength requirements. If the maximum stress is one hundred megapascals and the deformation is two millimeters in a certain iteration result, both are lower than the preset threshold, then the design scheme is confirmed to be feasible. This process reflects the iterative optimization characteristics of finite element analysis in engineering design, and finally a satisfactory design scheme is obtained through continuous adjustment and verification.

[0034] In step S105, the intelligent control system is used to adjust the injection molding parameters in real time within a single mold to achieve a continuous gradient transition of material properties and avoid poor interface bonding.

[0035] Current temperature distribution data in the mold is obtained, and an initial adjustment range of the injection molding parameters is determined in combination with a preset material property gradient curve. Pressure, temperature, and flow data during the injection molding process are collected in real time by the intelligent control system, and are compared and analyzed with the preset gradient property curve. If it is detected that the current material property distribution deviates from the preset curve, a dynamic adjustment amount of the injection molding parameters is calculated according to the degree of deviation. The injection pressure, injection speed, and mold temperature of the injection molding machine are adjusted in real time by the intelligent control system, so that the material property distribution gradually approaches the preset curve. The temperature change of the interface bonding area is continuously monitored to determine whether there is a temperature mutation or stress concentration phenomenon. If it is detected that the interface bonding area is abnormal, the injection molding parameters of the adjacent area are immediately adjusted to form a smooth temperature transition zone. After the injection molding is completed, the material property distribution data of the final molded part is obtained, and is compared with the preset gradient property curve to evaluate the interface bonding quality.

[0036] Specifically, the intelligent control system collects injection molding process data including material temperature, pressure, and flow. Taking injection molding of polyether ether ketone parts as an example, multiple temperature sensors are installed on the surface of the mold cavity to monitor the temperature changes in different areas in real time. By collecting the temperature data of the interface bonding area, it is found that there is a temperature mutation phenomenon at the interface of the two materials, and the temperature difference can reach 30°C, which will cause the interface bonding strength to decrease. By adjusting the injection molding parameters, a 20mm wide temperature transition zone is formed at the interface area, and the temperature difference is controlled within 10°C, effectively improving the interface bonding quality. The material performance gradient curve reflects the continuous change of material performance from one end of the product to the other end. Taking the elastic modulus as an example, the elastic modulus at one end of the product is required to reach 2500 megapascals, while the other end is required to drop to 1500 megapascals, and a smooth transition is needed in the middle area. By adjusting the injection molding parameters, the material produces different degrees of orientation during molding, thereby realizing the gradient change of performance. When the elastic modulus of a certain area deviates from the preset curve by more than 10%, the system will automatically adjust the injection pressure and speed of that area. The core of the intelligent control system is to establish the mapping relationship between the injection molding parameters and the material performance. Taking the injection pressure as an example, within the range of 20 megapascals to 40 megapascals, for every 5 megapascal increase in pressure, the crystallinity of the material will increase by about 3%, which directly affects the mechanical properties of the final product. The system judges whether the current pressure is within the target range according to the real-time collected pressure data, and if it deviates from the target value, it adjusts dynamically through a proportional valve. In terms of temperature control in the interface bonding area, a partitioned temperature control method is adopted. The surface of the mold cavity is divided into multiple temperature control areas, and the temperature of each area can be independently adjusted. Taking injection molding of a two-color cover plate as an example, the mold temperature of the hard area is set to 180°C, and the soft area is set to 160°C, and the interface is controlled by temperature gradient to ensure that the two materials are combined under good temperature conditions. When the temperature of the interface area is abnormal, the system will adjust the mold temperature of the adjacent area first to ensure uniform temperature distribution. The performance evaluation after injection molding adopts a non-destructive testing method. Through ultrasonic scanning technology, the internal structure information of the interface bonding area can be obtained. Combined with infrared thermal image analysis, the actual distribution of material performance can be determined. Taking the injection molding of a shock absorbing bracket as an example, it is found through scanning analysis that there are micro-pores in the interface bonding area, accounting for 5%, which exceeds the preset threshold of 3%. The system adjusts the injection molding process parameters to control the porosity within the allowable range, ensuring the product quality.

[0037] Step S106, during the injection molding process, the temperature and pressure changes in the mold are monitored in real time by sensors to obtain dynamic data of material flow and solidification.

[0038] The temperature value and the pressure value collected by the acquisition sensor in the mold are obtained to obtain real-time data; a corresponding relationship between a material flow state and a solidification state is analyzed according to a data change amount in the real-time data; a material state conversion feature is judged according to dynamic changes of the temperature value and the pressure value; if the temperature value reaches a preset threshold value, it is determined that the material enters a completion stage of the solidification state; if a preset abnormal fluctuation occurs in the pressure value, it is judged that the material flow state has a defect, and a defect position is determined; a complete state model of the material flow state and the solidification state is generated according to data distribution of a monitoring point; and process parameter settings in the mold are optimized according to the state model of the flow state and the solidification state.

[0039] Specifically, the mold sensor collects temperature and pressure data is the key link to obtain the dynamic information of injection molding process. In practical application, the temperature and pressure can be monitored in real time by arranging thermocouples and pressure sensors on the surface of the mold cavity. Taking the common polypropylene injection molding as an example, multiple monitoring points are arranged on the surface of the mold cavity, and temperature change data is collected during the injection molding process. If the measured temperature drops from two hundred and thirty degrees to one hundred and thirty degrees, it indicates that the material has begun to crystallize and solidify. For the judgment of material flow and solidification state, the temperature and pressure change curve characteristics can be analyzed. For example, during the injection filling stage, the pressure value will show a rapid rising trend, and when the pressure suddenly appears abnormal fluctuation, such as from sixty megapascals to forty megapascals, it indicates that there may be poor fusion or air pocket defects at this position. In the pressure maintaining stage, the temperature curve rate can reflect the solidification degree of the material, and when the temperature drop rate is lower than zero point five degrees per second, it indicates that the region has basically completed solidification. By establishing the state model of flow and solidification, the optimization adjustment of process parameters can be realized. By arranging multiple monitoring points in the mold cavity, the temperature field and pressure field distribution map are formed, and combined with the material flow path analysis, the potential defect area can be identified. For example, if the local area pressure value is low, it may cause sink marks or warping deformation, and at this time the pressure maintaining value or the pressure maintaining time of the region needs to be increased. If the temperature decreases too fast in a certain place, it may cause stress concentration, and the mold temperature should be appropriately increased or the cooling rate should be reduced. Different monitoring strategies can be adopted according to different product structure characteristics. For products with large wall thickness, the temperature change of the center area of the cavity should be focused on to ensure sufficient internal solidification. For thin-walled products, the pressure change of the flow front should be closely monitored to avoid excessive local pressure drop causing underfilling. Through comprehensive analysis of the monitoring data, process abnormalities can be found in time and adjusted to improve product quality stability. In actual production, a correlation model of process parameters and product quality can be established according to the collected historical data. For example, by analyzing the corresponding relationship between the temperature field distribution of different regions and the product size precision, a prediction model is established to guide the optimization of process parameters. When abnormal data is detected, the system automatically adjusts the relevant parameters, such as increasing or decreasing the local mold temperature, adjusting the injection speed, etc., to realize closed-loop control of product quality. This intelligent control method based on sensor data can effectively improve the stability of injection molding process and the product qualification rate.

[0040] Step S107, according to the dynamic data, whether the gradient distribution is consistent with the preset model, if not consistent, feedback to the intelligent control system, real-time optimization of injection molding parameters.

[0041] The dynamic data collected by the sensor in the injection molding process is obtained, and gradient distribution features are extracted for the dynamic data; the gradient distribution features are compared with a preset model to determine whether the gradient distribution features deviate from the preset model; if the gradient distribution features deviate from the preset model, an error value is extracted and an optimization amount is calculated; the injection parameters of the intelligent control system are adjusted according to the optimization amount to generate a new parameter combination; a regression model in a machine learning algorithm is used to predict the effect corresponding to the new parameter combination to obtain a prediction result; if the prediction result meets a preset condition, the parameter configuration of the intelligent control system is updated; and the optimized parameter configuration and the dynamic data are recorded by a data storage module.

[0042] Specifically, in the injection molding process, the dynamic data collected by the sensor includes key process parameters such as mold wall temperature, injection pressure, and holding pressure. These data show different gradient distribution features, such as the temperature gradient in the flow channel gradually decreases from the inlet to the end, and the pressure gradient shows a trend of first increasing and then decreasing. The preset model is usually established based on the ideal state under standard process conditions, such as the temperature gradient should be kept at two to three degrees per centimeter, and the pressure gradient should show a smooth transition. When the gradient distribution deviates in actual production, specific deviation features need to be extracted. For example, in the production process of a certain injection molded part, the temperature gradient of a certain area of the mold suddenly increases to five degrees per centimeter, far exceeding the preset three-degree standard, and the pressure gradient of the area also shows a sharp fluctuation. In this case, the system calculates the optimization amount of temperature and pressure, such as reducing the mold temperature setting value of the area by ten degrees and appropriately increasing the holding pressure value. The intelligent control system automatically adjusts the parameters according to the optimization amount to form a new parameter combination. For example, the mold temperature of the area is adjusted from one hundred and eighty degrees to one hundred and seventy degrees, the holding pressure is increased from forty megapascals to forty-five megapascals, and the injection speed is reduced from fifty millimeters per second to forty-five millimeters per second. The coordinated adjustment of these parameters can help stabilize the molding process. When the regression model predicts the effect of the new parameters, it will consider the mutual influence of multiple process indicators. For example, a decrease in mold temperature will lead to a decrease in flowability, so it is necessary to compensate by increasing the holding pressure to ensure the filling density of the product. The prediction model will calculate the product quality indicators such as warpage deformation and sink mark depth based on historical data under this parameter combination. After meeting the preset conditions, the system will update the parameter configuration and store it. For example, after optimization, the quality indicators of a certain injection molded product meet the requirements: the warpage deformation is controlled within zero point two millimeters, the sink mark depth is less than zero point one millimeter, and the product size tolerance is kept within a range of plus or minus zero point zero five millimeters. These data and the corresponding process parameters will be recorded for subsequent process optimization and quality traceability. In this way, a complete data closed loop is formed to continuously improve the level of injection molding process.

[0043] Step S108, using a machine learning algorithm, training a performance prediction model based on historical injection data, optimizing the initial setting of the material distribution parameters, and improving production efficiency.

[0044] Obtain historical injection data, the historical injection data including material parameters and distribution parameters; based on the historical injection data, train a performance prediction model using a random forest algorithm; predict the production efficiency under different combinations of material parameters and distribution parameters through the performance prediction model to obtain the predicted production efficiency; determine whether the predicted production efficiency is lower than a preset threshold, if so, optimize the initial setting of the material parameters and distribution parameters using a support vector machine algorithm to obtain optimized material parameters and distribution parameters; retrain the performance prediction model according to the optimized material parameters and distribution parameters to obtain a retrained performance prediction model; predict the optimized production efficiency through the retrained performance prediction model; and determine whether the optimized production efficiency reaches an expected target.

[0045] Specifically, in injection molding production, the acquisition of historical data includes material parameters such as the melting point and flowability of plastic particles, as well as distribution parameters such as temperature distribution and pressure distribution. Through real-time data collected by various sensors during the production process, a complete historical data set can be formed. For example, the flowability of polypropylene material at different temperatures, the pressure distribution in each area of the mold, and other key indicators. The random forest algorithm predicts performance by building multiple decision trees, each trained on a different subset of features. For example, set the mold temperature to two hundred and thirty degrees, the injection pressure to eighty megapascals, and the holding time to ten seconds, and predict the product pass rate and production cycle under these conditions. By analyzing a large amount of historical data, the algorithm can identify key parameter combinations that affect production efficiency. When the predicted production efficiency is lower than the target value, the initial parameter settings need to be adjusted. For example, the original screw speed is set to sixty revolutions per minute, but the predicted production capacity is only eighty pieces per hour, which is lower than the target value of one hundred pieces per hour. At this time, the parameter settings need to be optimized. The support vector machine algorithm can find better parameter combinations by establishing an optimal classification hyperplane. For example, increase the screw speed to seventy revolutions per minute and appropriately increase the barrel temperature. After optimizing the parameters, the performance prediction model needs to be retrained. The new training data includes the optimized parameter combinations and their corresponding production performance. In this way, the model can more accurately predict the production efficiency under different parameter settings. For example, the original model prediction error is around ten percent, and after retraining, the error can be reduced to within five percent. This dynamic optimization process ensures that the model can always accurately reflect the current production situation. For different plastic raw materials, the optimization strategy also needs to be adjusted accordingly. For example, polycarbonate is sensitive to temperature, so it needs to focus on optimizing temperature distribution parameters. For materials such as polyethylene, more attention needs to be paid to the optimization of pressure distribution parameters. By establishing a mapping relationship between material characteristics and process parameters, parameter optimization can be more targeted. In practical applications, the influence of mold structure on parameter optimization also needs to be considered. For example, for thin-walled products, the flow path is longer, which requires special attention to the uniformity of pressure distribution during optimization. For thick-walled products, it is necessary to focus on optimizing the holding parameters to prevent shrinkage and warping defects. Through this multi-dimensional parameter optimization, the stability and efficiency of injection molding production can be significantly improved.

[0046] Step S109, through the automatic detection equipment, the heat dissipation performance and structural strength of the shell after molding are tested to ensure the consistency and yield of the product.

[0047] The temperature distribution data and stress distribution data of the formed shell are acquired to obtain a first detection data set; the first detection data set is preprocessed to remove abnormal values and noise data, to generate a second detection data set; a performance index matrix is constructed by extracting heat dissipation performance indexes and structural strength indexes from the second detection data set; whether each index in the performance index matrix meets the standard is judged according to a preset performance threshold, to obtain a preliminary judgment result; the preliminary judgment result is optimized by using a machine learning algorithm to correct misjudgments and omissions, to obtain a final judgment result; a yield rate is calculated according to the final judgment result; and the yield rate and the performance index matrix are analyzed to establish a quality control model.

[0048] Specifically, the automated detection equipment includes an infrared thermal imager and a stress and strain detector. The thermal imager can monitor the surface temperature distribution of the formed shell in real time, and the stress and strain detector measures the stress distribution of each key point of the shell through a sensor array. For example, a plurality of detection points are arranged on the surface of the shell, and temperature and stress data are collected at each point to form a multi-dimensional data matrix. The data preprocessing link focuses on identifying and removing abnormal data points. In the temperature data, if the temperature of a certain point suddenly drops or rises more than three times the average value of adjacent points, it is determined to be abnormal. If the stress data deviates from the normal distribution and appears as an outlier, it also needs to be removed. Through such preprocessing, the data quality for subsequent analysis can be ensured. The performance index matrix includes two categories of indexes, namely heat dissipation performance and structural strength. The heat dissipation performance mainly investigates temperature uniformity and the highest temperature point, such as a temperature difference of no more than ten degrees on the surface of the shell and a maximum temperature of no more than seventy degrees. The structural strength indexes include a stress concentration coefficient and a deformation amount, such as a stress concentration coefficient of less than two and a maximum deformation amount of no more than 0.5 mm. The preliminary judgment uses a threshold method to compare each index with a preset standard. For example, if the temperature uniformity is out of standard but the structural strength is qualified, it is determined that the heat dissipation performance is not up to standard. The machine learning algorithm can use a support vector machine for optimization, establish a classification model through training data, and improve the accuracy of judgment. The yield rate is calculated based on the final judgment result. For example, if ninety-five out of one hundred continuously produced products are qualified, the yield rate is ninety-five percent. The quality control model finds out the key factors affecting product quality by analyzing the correlation between the yield rate and each performance index. For example, if it is found that the temperature uniformity is strongly correlated with the yield rate, the temperature control parameters can be optimized. Through the closed-loop management of data acquisition, analysis, judgment and optimization, continuous improvement of the production process is realized. For example, if it is found that the yield rate of a certain batch decreases, it is found through analysis that it is caused by uneven mold temperature, and the heating parameters are adjusted in time to restore the product quality to the normal level. This intelligent management method based on data can effectively improve the stability of product quality.

[0049] The above-described embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A gradient injection molding method of a Bluetooth earphone case shell, characterized in that, The method comprises the following steps: S101, using a preset gradient material database, obtaining thermal conductivity and mechanical strength data of the material, and establishing a relationship model between material performance and gradient distribution; The S101 further comprises: obtaining a preset gradient material database, wherein the gradient material database comprises thermal conductivity and mechanical strength data of the material; obtaining gradient distribution parameters of the material, calculating gradient distribution characteristics of the material according to the gradient distribution parameters; using a first machine learning algorithm to establish a first mapping relationship between the thermal conductivity and the gradient distribution characteristics; using a second machine learning algorithm to establish a second mapping relationship between the mechanical strength and the gradient distribution characteristics; if the thermal conductivity and the mechanical strength have a correlation, obtaining correlation data of the thermal conductivity and the mechanical strength, generating a comprehensive performance model of the material according to the correlation data; determining the relationship between the material performance and the gradient distribution characteristics according to the comprehensive performance model; obtaining target gradient distribution parameters, calculating target gradient distribution characteristics according to the target gradient distribution parameters; determining performance parameters of the material under the target gradient distribution characteristics according to the relationship between the material performance and the gradient distribution characteristics; S102, determining initial parameters of material distribution in the gradient injection molding process according to the relationship model, including setting values of temperature, pressure and injection speed; The S102 further comprises: obtaining historical data of material distribution in the gradient injection molding process, establishing a correlation relationship model of the temperature, the pressure and the injection speed; calculating the best temperature value, pressure value and speed value range of the material distribution according to the correlation relationship model; determining the initial setting values of the temperature, the pressure and the injection speed according to the best temperature value, pressure value and speed value range; obtaining material distribution state information, and judging whether the material distribution state is abnormal; if the material distribution state is abnormal, adjusting the setting values of the temperature, the pressure and the injection speed, and recalculating the material distribution state; monitoring the material distribution state in real time, and judging whether the setting values of the temperature, the pressure and the injection speed reach a predetermined target; if the setting values of the temperature, the pressure and the injection speed do not reach the predetermined target, automatically adjusting the setting values of the temperature, the pressure and the injection speed, and optimizing the material distribution state; using a machine learning algorithm to learn the adjusted setting values of the temperature, the pressure and the injection speed, and updating the correlation relationship of the temperature, the pressure and the injection speed in the correlation relationship model; S103, simulating the flow and solidification process of the material in the mold by using a finite element analysis tool, and predicting the thermal conductivity performance and mechanical strength of the gradient distribution; The S103 further comprises: establishing a mathematical model of material flow and solidification by using a finite element tool, obtaining flow state and solidification state data of the material in the mold; calculating the heat conduction performance value and the mechanical strength value inside the material according to the flow state and the solidification state data through a gradient distribution algorithm; predicting the gradient distribution map of the material in the mold by using a regression model in the machine learning algorithm for the heat conduction performance value and the mechanical strength value; if the predicted gradient distribution map matches a preset threshold range, it is determined that the performance value of the material meets the design requirement; if not, adjusting the simulation parameters of the flow state and the solidification state, recalculating the heat conduction performance value and the mechanical strength value; according to the adjusted parameters, simulating the flow and solidification process of the material in the mold again by using the finite element tool, obtaining new flow state and solidification state data; using the new flow state and solidification state data to regenerate the gradient distribution map, and judging whether the material performance value meets the design requirement; S104, if the simulation result does not reach the preset performance threshold value, adjusting the material distribution parameters and re-performing the finite element analysis until the heat dissipation and structural strength requirements are met; The S104 further comprises: obtaining a result value of the finite element analysis; comparing the result value with a preset value to obtain a comparison result; if the comparison result shows that the result value does not reach the preset value, calculating an adjustment amount and adjusting the material distribution parameters according to the adjustment amount to obtain adjusted material distribution parameters; re-performing the finite element analysis according to the adjusted material distribution parameters to obtain a new result value; judging whether the new result value reaches the preset value; if the new result value reaches the preset value, outputting the material distribution parameters that meet the heat dissipation and structural strength; if the new result value does not reach the preset value, repeating the steps of adjusting the material distribution parameters and re-performing the finite element analysis and judging whether the new result value reaches the preset value until the new result value reaches the preset value; S105, using an intelligent control system to adjust the injection molding parameters in a single mold in real time, realizing continuous gradient transition of the material performance, and avoiding the problem of poor interface bonding; The S105 further comprises: obtaining current temperature distribution data in the mold, combining a preset material performance gradient curve to determine an initial adjustment range of injection molding parameters; through the intelligent control system, real-time acquisition of pressure, temperature and flow data in the injection molding process is performed, and comparison and analysis with the preset gradient performance curve is performed, if it is detected that the current material performance distribution deviates from the preset curve, then a dynamic adjustment amount of the injection molding parameters is calculated according to the deviation degree; the intelligent control system is used to real-time adjust the injection pressure, injection speed and mold temperature of the injection molding machine, so that the material performance distribution gradually approaches the preset curve; the temperature change of the interface bonding area is continuously monitored to determine whether there is a temperature mutation or stress concentration phenomenon, if it is detected that the interface bonding area appears abnormity, then the injection molding parameters of the adjacent area are immediately adjusted to form a smooth temperature transition zone; after the injection molding is completed, the material performance distribution data of the final molded part is obtained, and comparison with the preset gradient performance curve is performed to evaluate the interface bonding quality; S106, in the injection molding process, the temperature and pressure changes in the mold are monitored in real time through the sensor to obtain dynamic data of material flow and solidification; The S106 further comprises: obtaining temperature and pressure values collected by the sensor in the mold to obtain real-time data; for the data change amount in the real-time data, the corresponding relationship between the material flow state and the solidification state is analyzed; the material state conversion characteristics are judged according to the dynamic changes of the temperature and pressure values; if the temperature value reaches a preset threshold, it is determined that the material enters the completion stage of the solidification state; if the pressure value appears a preset abnormal fluctuation, it is judged that the material flow state has defects, and the defect position is determined; according to the data distribution of the monitoring points, a complete state model of the material flow state and the solidification state is generated; according to the state model of the flow state and the solidification state, the process parameter setting in the mold is optimized; S107, according to the dynamic data, whether the gradient distribution is consistent with the preset model is judged, if not, it is fed back to the intelligent control system to real-time optimize the injection molding parameters; S108, using a machine learning algorithm, a performance prediction model is trained based on historical injection data to optimize the initial setting of the material distribution parameters and improve the production efficiency; S109, through an automatic detection device, the heat dissipation performance and structural strength of the molded shell are tested to ensure the consistency and yield of the product; The S109 further comprises: obtaining temperature distribution data and stress distribution data of the molded shell to obtain a first detection data set; preprocessing the first detection data set to remove abnormal values and noise data to generate a second detection data set; extracting heat dissipation performance indicators and structural strength indicators from the second detection data set to construct a performance indicator matrix; according to a preset performance threshold, it is judged whether each indicator in the performance indicator matrix meets the standard to obtain a preliminary judgment result; using a machine learning algorithm, the preliminary judgment result is optimized to correct false negatives and omissions to obtain a final judgment result; according to the final judgment result, the yield is calculated; the yield and the performance indicator matrix are analyzed to establish a quality control model.

2. The method of claim 1, wherein, The S107 comprises: acquiring dynamic data collected by a sensor in an injection molding process, extracting gradient distribution features for the dynamic data; comparing the gradient distribution features with a preset model to determine whether the gradient distribution features deviate from the preset model; if the gradient distribution features deviate from the preset model, extracting a deviation value and calculating an optimization amount; adjusting injection molding parameters of an intelligent control system according to the optimization amount to generate a new parameter combination; using a regression model in a machine learning algorithm to predict an effect corresponding to the new parameter combination to obtain a prediction result; if the prediction result meets a preset condition, updating parameter configurations of the intelligent control system; recording the optimized parameter configurations and the dynamic data through a data storage module.

3. The method of claim 1, wherein, The S108 comprises: acquiring historical injection molding data, the historical injection molding data including material parameters and distribution parameters; based on the historical injection molding data, training a performance prediction model using a random forest algorithm; predicting production efficiency under different combinations of material parameters and distribution parameters through the performance prediction model to obtain a predicted production efficiency; determining whether the predicted production efficiency is lower than a preset threshold, if yes, optimizing initial settings of the material parameters and the distribution parameters using a support vector machine algorithm to obtain optimized material parameters and distribution parameters; retraining the performance prediction model according to the optimized material parameters and the distribution parameters to obtain a retrained performance prediction model; predicting an optimized production efficiency through the retrained performance prediction model; determining whether the optimized production efficiency reaches an expected target.

Images