A process control method and system for composite material forming

Abstract

The present invention discloses a process control method and system for composite material forming, and the present invention relates to the technical field of composite material forming control. The method includes the following steps: obtaining composite material finished products produced with different types and different addition amounts of fillers, obtaining the thermal conductivity of the finished products through experimental detection, and generating a training sample data set by mapping the filler type and addition amount to the thermal conductivity. Based on this data set, a neural network prediction model is established, with the filler type and addition amount used as inputs and the thermal conductivity used as a label for training. Subsequently, the filler information of the finished product to be produced is input to obtain a predicted value of the thermal conductivity. Combining the real-time process flow parameters and the curing degree of the finished product, the curing degree is corrected and the heat demand coefficient, resin reaction degree coefficient, and pressure loss coefficient are calculated. Finally, according to these coefficients, the process control parameters are dynamically corrected to achieve precise control of the composite material forming process. The efficiency of composite material forming and the quality of the finished products are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of composite material forming control, and specifically to a process control method and system for composite material forming. Background Art

[0002] In the production of modern composite materials, with the increasing requirements for material properties, especially in terms of thermal conductivity, strength, corrosion resistance, etc., the forming process control of composite materials has become increasingly complex. Traditional composite material forming processes usually rely on experience and experimental data for parameter optimization. This method is not only time-consuming and laborious, but also difficult to meet the diverse requirements of material properties. The selection of fillers has a significant impact on the final properties of composite materials, and different types and different addition amounts of fillers may lead to significant differences in the thermal conductivity of the finished products during the production process. This makes the development of an effective process control method an urgent technical problem to be solved.

[0003] Existing technologies often lack a systematic study on the relationship between filler types and addition amounts and the properties of composite materials, resulting in the inability to accurately predict the thermal conductivity of materials during the production process. Due to the lack of effective control means, the thermal conductivity may not meet the expectations during the processing, thus affecting the quality and performance of the finished products. In addition, process parameters such as resin viscosity, heat flux density, and environmental humidity during the forming process also have an important impact on the final properties of the finished products, and these parameters are often difficult to monitor and adjust in real time during the production process.

[0004] Therefore, there is an urgent need for a new composite material forming process control method. By introducing advanced prediction models and dynamic correction mechanisms, precise control of the properties of composite materials can be achieved. Using advanced machine learning technologies such as neural networks, a mapping relationship between filler types, addition amounts, and the thermal conductivity of the finished products is established, so as to accurately predict the thermal conductivity of the finished products. At the same time, the process parameters are monitored and adjusted in real time to ensure that flexible responses can be made according to the actual situation during the forming process, improving the overall performance and production efficiency of composite materials.

[0005] In the prior art, the published number CN118664933A discloses a process control method and device for composite material forming. The method includes: obtaining the initial parameters of the composite material; establishing a mathematical model of the composite material forming process according to the initial parameters; based on the mathematical model, performing numerical simulation on the composite material forming process to obtain a numerical simulation result; comparing the numerical simulation result with a preset forming result, screening out the optimal process parameter combination that meets the preset forming result as the composite material forming control parameter; setting parameters for the temperature control system and pressure control system of the composite material forming equipment, and starting the forming process of the composite material to complete the production of the composite material product. However, in this method, the forming process of the composite material involves complex physical and chemical reactions. Establishing an accurate mathematical model not only requires profound material science knowledge but also needs to fully consider various factors in the forming process, including temperature, pressure, flow characteristics, etc. If the model cannot accurately capture the relationship between these factors, it may lead to deviations in the prediction results, thereby affecting the selection of process parameters. At the same time, during the production process, the product performance and process parameters may change with time and conditions. This method fails to mention how to monitor the finished product performance in real time and make adjustments according to the actual feedback. This lack of real-time feedback control method may lead to low production efficiency and unstable finished product quality. Therefore, this method reduces the accuracy and effectiveness of process parameter control.

[0006] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present disclosure. Therefore, it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0007] The purpose of the present invention is to provide a process control method and system for composite material forming to solve the problems raised in the above background art.

[0008] To achieve the above purpose, the present invention provides the following technical solutions:

[0009] A process control method for composite material forming, the specific steps include:

[0010] Obtain composite material finished products produced with several different types and different addition amounts of fillers, detect the composite material finished products through experiments, obtain the corresponding thermal conductivity of the finished products, map the type and addition amount of the fillers to the corresponding thermal conductivity of the finished products one by one, and generate a training sample data set;

[0011] Based on the data in the training sample data set, establish a neural network prediction model, use the type and addition amount of the fillers in the training sample data set as the input of the model, and use the corresponding thermal conductivity of the finished products as the label to train the neural network prediction model to obtain a thermal conductivity prediction model;

[0012] Obtain the type and addition amount of the filler added to the composite material finished product to be produced, input the type and addition amount of the filler added to the composite material finished product to be produced into the trained thermal conductivity prediction model, obtain the predicted value of the thermal conductivity of the finished product to be produced, and at the same time obtain the real-time process flow parameters during the processing, where the process flow parameters include resin viscosity, heat flux density, environmental humidity, resin injection speed, and air flow rate;

[0013] Obtain the curing degree of the finished product in real time during the processing, correct the curing degree of the finished product based on the process flow parameters to obtain the accurate value of the curing degree of the finished product, and calculate the heat demand coefficient, resin reaction degree coefficient, and pressure loss coefficient according to the accurate value of the curing degree of the finished product, the predicted value of the thermal conductivity, and the process flow parameters;

[0014] Dynamically correct the process control parameters during the processing according to the obtained heat demand coefficient, resin reaction degree coefficient, and pressure loss coefficient to obtain the real-time correction value of the process control parameters, and control the processing based on the obtained real-time correction value of the process control parameters to complete the process control of the composite material forming, where the process control parameters include mold temperature, curing pressure, and curing time.

[0015] Further, the generation method of the training sample data set is as follows: Map the type and addition amount of the filler one-to-one with the finished product thermal conductivity parameters of the corresponding composite material finished product to form a corresponding grid, and record the formed grid as the training sample data set;

[0016] Based on the convolutional neural network, establish a thermal conductivity prediction model, where the convolutional neural network consists of an input layer, a convolutional layer, a pooling layer, a fully connected layer, and an output layer, and the activation function in the convolutional layer is function, The specific expression of the function is:

[0017]

[0018] Among them, represents the th corresponding convolutional layer, then represents the th feature value of the rd feature data in the th corresponding convolutional layer, where is the index of the convolutional layer, is the index of the feature data, is the index of the feature value in the feature data, where , , , where N, M, and O are the total numbers of the convolutional layer, feature data, and feature values in the feature data respectively;

[0019] For the fully connected layer, set the number of neurons in the fully connected layer to 32, set the initial learning rate of the neural network to 0.001, and set the number of training epochs to 100;

[0020] The input of the trained thermal conductivity prediction model is the type and addition amount of the filler, and the output is the predicted value of the finished product thermal conductivity of the composite material.

[0021] Furthermore, obtain the real-time degree of cure of the finished product during the processing, correct the degree of cure of the finished product based on the process flow parameters to obtain the accurate value of the degree of cure of the finished product, where the formula for calculating the accurate value of the degree of cure of the finished product is:

[0022]

[0023] In the formula, is the accurate value of the degree of cure of the finished product at the current moment, is the degree of cure of the finished product collected at the current moment, is the initial resin viscosity, is the resin viscosity at the current moment, is the resin viscosity influence constant;

[0024] where the degree of cure of the finished product at the current moment is specifically the degree of cure monitored in real time by techniques such as infrared spectroscopy or thermogravimetric analysis.

[0025] Furthermore, according to the accurate value of the degree of cure of the finished product, the predicted value of the thermal conductivity, and the process flow parameters, calculate the heat demand coefficient, the resin reaction degree coefficient, and the pressure loss coefficient, where the formula for calculating the heat demand coefficient is:

[0026]

[0027] In the formula, is the heat demand coefficient at the current moment, is the air flow rate of the processing environment at the current moment, is the cumulative reaction heat at the current moment, is the predicted value of the finished product thermal conductivity;

[0028] where the cumulative reaction heat at the current moment is calculated through the heat flux density, where the specific calculation formula is:

[0029]

[0030] In the formula, is the initial reaction heat, is the start time of the processing, is the current moment, is the heat flux density, and the heat flux density is collected in real time by a heat flow meter.

[0031] Further, the formula based on which the resin reaction degree coefficient is calculated is:

[0032]

[0033] In the formula, is the resin reaction degree coefficient at the current moment, is the environmental humidity;

[0034] The formula based on which the pressure loss coefficient is calculated is:

[0035]

[0036] In the formula, is the pressure loss coefficient of the injected resin at the current moment, is the length of the resin injection runner, is the diameter of the resin injection runner, is the density of the resin, is the resin injection speed of the resin, is the friction factor.

[0037] Further, the process control parameters in the processing process are dynamically corrected according to the obtained heat demand coefficient, resin reaction degree coefficient and pressure loss coefficient to obtain the real-time correction value of the process control parameters. The formula specifically based on which the die temperature correction value is calculated is:

[0038]

[0039] In the formula, is the die temperature correction value at the current moment, is the initial die temperature set, is the heat demand coefficient at the current moment;

[0040] The formula specifically based on which the curing time correction value is calculated is:

[0041]

[0042] In the formula, is the curing time correction value at the current moment, is the initial curing time set;

[0043] The formula specifically based on which the curing pressure correction value is calculated is:

[0044]

[0045] In the formula, is the curing pressure correction value at the current moment, is the initial value of the set curing pressure, where , and are the weight constants of the heat demand coefficient, the resin reaction degree coefficient, and the pressure loss coefficient respectively, where and , and are all greater than 0.

[0046] The present invention also provides a process control system for composite material molding. The process control system for composite material molding is used to execute the above-mentioned process control method for composite material molding, and includes:

[0047] A sample data acquisition module, which is used to obtain composite material finished products produced with several different types and different addition amounts of fillers, detect the composite material finished products through experiments, obtain the corresponding thermal conductivity of the finished products, map the type and addition amount of the fillers to the corresponding thermal conductivity of the finished products one by one, and generate a training sample data set;

[0048] A prediction model training module, which is used to establish a neural network prediction model based on the data in the training sample data set, use the type and addition amount of the fillers in the training sample data set as the input of the model, and use the corresponding thermal conductivity of the finished products as labels to train the neural network prediction model to obtain a thermal conductivity prediction model;

[0049] A process parameter analysis module, which is used to obtain the type and addition amount of the fillers added to the composite material finished product to be produced, input the type and addition amount of the fillers added to the composite material finished product to be produced into the trained thermal conductivity prediction model to obtain the predicted value of the thermal conductivity of the finished product to be produced, and at the same time obtain the real-time process flow parameters during the processing. The process flow parameters include resin viscosity, heat flux density, environmental humidity, resin injection speed, and air flow rate;

[0050] A curing degree detection module, which is used to obtain the real-time curing degree of the finished product during the processing, correct the curing degree of the finished product based on the process flow parameters to obtain the accurate value of the curing degree of the finished product, and calculate the heat demand coefficient, the resin reaction degree coefficient, and the pressure loss coefficient according to the accurate value of the curing degree of the finished product, the predicted value of the thermal conductivity, and the process flow parameters;

[0051] A control parameter correction module, which is used to dynamically correct the process control parameters during the processing according to the obtained heat demand coefficient, resin reaction degree coefficient, and pressure loss coefficient to obtain the real-time correction value of the process control parameters, and control the processing process based on the obtained real-time correction value of the process control parameters to complete the process control of composite material molding. The process control parameters include mold temperature, curing pressure, and curing time.

[0052] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0053] Firstly, by creating a training sample data set based on the correspondence between the filler type and addition amount and the thermal conductivity of the finished product, and using a neural network for prediction modeling, the thermal conductivity of the composite material to be produced can be predicted quickly and accurately. This improves the predictability of production, provides a scientific basis for material selection and proportioning, and reduces material waste and production risks caused by mismatching. Secondly, this method combines the process parameters monitored in real time with the degree of cure of the finished product for dynamic correction, ensuring a high degree of consistency between the performance of the finished product and the production conditions. By monitoring and adjusting parameters such as resin viscosity, heat flux density, and environmental humidity during the processing, the process control parameters such as mold temperature, curing pressure, and curing time can be optimized in real time. This flexible dynamic adjustment mechanism effectively enhances the adaptability of the processing process and ensures that the composite material can meet the expected performance standards under different production conditions. Finally, the implementation of this method not only improves the quality and reliability of the finished product, but also significantly reduces the production cost and time. By reducing unnecessary test and correction time, the production process is optimized. Brief Description of the Drawings

[0054] Figure 1 It is a schematic diagram of the overall method flow of the present invention;

[0055] Figure 2 It is a schematic diagram of the overall system structure of the present invention. Detailed Embodiments

[0056] To make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the following further elaborates on the present invention in conjunction with specific embodiments.

[0057] It should be noted that unless otherwise defined, the technical terms or scientific terms used in the present invention should have the ordinary meanings understood by those with ordinary skills in the field to which the present invention pertains. The "first", "second", and similar terms used in the present invention do not denote any order, quantity, or importance, but are only used to distinguish different components. The terms such as "comprising" or "including" mean that the elements or objects appearing before this term cover the elements or objects listed after this term and their equivalents, without excluding other elements or objects. The terms such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The terms such as "upper", "lower", "left", "right", etc. are only used to represent relative positional relationships, and when the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0058] Embodiment:

[0059] Please refer toFigure 1 , the present invention provides a technical solution:

[0060] A process control method for composite material forming, the specific steps include:

[0061] Step 1: Obtain composite material finished products produced with several different types and different addition amounts of fillers, detect the composite material finished products through experiments, obtain the corresponding thermal conductivity of the finished products, map the type and addition amount of the fillers one by one to the corresponding thermal conductivity of the finished products, and generate a training sample data set.

[0062] In the production of composite materials, the type and addition amount of fillers have an important impact on the thermal conductivity of the finished products. Fillers are materials used to improve the performance of composite materials. Common types of fillers include: Inorganic fillers: such as alumina, silicate, graphite, calcium carbonate, etc. These fillers usually have good thermal conductivity and chemical stability; Organic fillers: such as polymer particles, glass fibers, etc. These fillers usually have less impact on thermal conductivity, but can improve the toughness of the material; Thermal conductive fillers: such as metal powders (copper powder, aluminum powder), thermal conductive plastics, etc. These specially designed fillers have high thermal conductivity and can significantly improve the thermal conductivity of composite materials; Nano fillers: such as carbon nanotubes, graphene, etc. These fillers are widely studied and applied in high-performance composite materials due to their high specific surface area and excellent thermal conductivity.

[0063] The addition amount of fillers is usually expressed in mass percentage (%). Different addition amounts have a significant impact on the final performance of composite materials; for example: Low addition amount (1% - 5%): May be mainly used to improve the physical properties of the material, such as toughness, wear resistance, etc., and have less impact on thermal conductivity; Medium addition amount (5% - 15%): In this range, the thermal conductivity of the filler will gradually appear, and can effectively enhance the thermal conductivity of the composite material; High addition amount (above 15%): The filler may form continuous thermal conduction channels, greatly improving the thermal conductivity, but may also affect other properties, such as mechanical strength and toughness.

[0064] The thermal conductivity of the filler itself is the main factor affecting the thermal conductivity of composite materials. For example, the thermal conductivity of copper powder is higher than that of graphite powder, so adding copper powder will significantly increase the thermal conductivity. There may be interactions between different types of fillers, affecting the thermal conductivity. For example, adding graphite filler can improve the thermal conductivity of the polymer matrix, etc.

[0065] The method for generating the training sample data set is: Map the type and addition amount of the fillers one by one to the thermal conductivity parameters of the corresponding composite material finished products to form a corresponding grid, and record the formed grid as the training sample data set.

[0066] Step 2: Based on the data in the training sample dataset, establish a neural network prediction model. Use the type and addition amount of the filler in the training sample dataset as the input of the model, and use the corresponding finished product thermal conductivity as the label to train the neural network prediction model to obtain a thermal conductivity prediction model.

[0067] Based on a convolutional neural network, establish a thermal conductivity prediction model, where the convolutional neural network consists of an input layer, a convolutional layer, a pooling layer, a fully connected layer, and an output layer. The activation function in the convolutional layer is function, The specific expression of the function is:

[0068]

[0069] where, represents the th corresponding convolutional layer, represents the th corresponding convolutional layer, and the th eigenvalue of the th feature data. Among them, is the index of the convolutional layer, is the index of the feature data, is the index of the eigenvalue in the feature data. Among them, , , , where N, M, and O are the total numbers of convolutional layers, feature data, and eigenvalues in the feature data, respectively;

[0070] For the fully connected layer, set the number of neurons in the fully connected layer to 32, set the initial neural network learning rate to 0.001, and set the number of training epochs to 100;

[0071] The input of the trained thermal conductivity prediction model is the type and addition amount of the filler, and the output is the predicted value of the finished product thermal conductivity of the composite material.

[0072] One of the greatest advantages of a convolutional neural network (CNN) is its ability to automatically extract features from input data. When establishing a thermal conductivity prediction model, the type and addition amount of the filler are used as input data, and the CNN can automatically identify and extract potential features related to thermal conductivity without manual feature extraction. This is especially important because the type and addition amount of the filler may contain complex non-linear relationships, and manual extraction of these features is usually very difficult and error-prone.

[0073] The relationship between thermal conductivity and the type and addition amount of the filler may be non-linear. Traditional linear regression models may not be able to accurately capture such complex non-linear relationships. Through its multi-layer structure, the CNN can learn such complex non-linear relationships, thereby improving the prediction ability of the model.

[0074] A CNN can learn high - order features of the input data through multiple convolutional layers, and the pooling layer helps reduce the dimension and extract the most representative features. In this way, the CNN can capture the deep - level features affecting thermal conductivity from the original inputs such as filler type and addition amount.

[0075] Step 3: Obtain the type and addition amount of the filler added to the composite material product to be produced, input the type and addition amount of the filler added to the composite material product to be produced into the trained thermal conductivity prediction model to obtain the predicted value of the thermal conductivity of the product to be produced. At the same time, obtain the real - time process flow parameters during the processing, and the process flow parameters include resin viscosity, heat flux density, environmental humidity, resin injection speed, and air flow rate.

[0076] Resin viscosity is an important parameter affecting the molding of composite materials and can be obtained by the following methods: Using an on - line viscosity monitoring system (such as an ultrasonic viscometer or an optical viscometer), these devices can be installed in the resin conveying pipeline to monitor viscosity changes in real time; or using a dedicated instrument such as a rotational viscometer or a capillary viscometer to directly measure the resin viscosity. The rotational viscometer measures the resistance of the fluid by rotating a propeller or a rod and then calculates the viscosity.

[0077] Heat flux density describes the heat passing through per unit area and is often obtained by the following methods: Heat flux meter: Using a heat flux sensor (such as a thermocouple or a heat flux plate) to measure the heat flux density at various positions during the molding process of the composite material. The heat flux meter is usually embedded in the mold and can monitor temperature changes in real time; Infrared thermal imager: Using an infrared thermal imager, the temperature distribution on the surface of the mold can be obtained non - contactively, and the heat flux density can be calculated through a heat conduction model.

[0078] Environmental humidity has an important impact on material properties and can be obtained by the following methods: Humidity sensor: Using a digital or analog humidity sensor (such as a capacitive humidity sensor or an impedance - type humidity sensor), these sensors can monitor the relative humidity in the air in real time.

[0079] The resin injection speed directly affects the molding process and can be obtained by the following methods:

[0080] Using a flow meter (such as an electromagnetic flow meter or an ultrasonic flow meter) to measure the flow rate of the resin. The flow meter can monitor the fluid flow in real time and calculate the injection speed through the volume change of the fluid, or installing a position sensor or a speed sensor in the injection system to monitor the working speed of the syringe to obtain the actual injection speed.

[0081] The air flow rate usually has an impact on the forming process and is commonly obtained by the following methods: using an anemometer (such as a hot-wire anemometer or a rotating anemometer) to measure the air flow velocity. The anemometer can provide real-time air flow rate data; or using a differential pressure sensor in the air flow pipeline to monitor the pressure change of the air flow and calculate the air flow rate based on the principle of fluid dynamics.

[0082] Step 4: Obtain the real-time curing degree of the finished product during the processing, correct the curing degree of the finished product based on the process parameters to obtain the accurate value of the curing degree of the finished product, and calculate the heat demand coefficient, resin reaction degree coefficient, and pressure loss coefficient according to the accurate value of the curing degree of the finished product, the predicted value of the thermal conductivity, and the process parameters.

[0083] Obtain the real-time curing degree of the finished product during the processing, correct the curing degree of the finished product based on the process parameters to obtain the accurate value of the curing degree of the finished product, where the formula for calculating the accurate value of the curing degree of the finished product is:

[0084]

[0085] In the formula, is the accurate value of the curing degree of the finished product at the current moment, is the curing degree of the finished product collected at the current moment, is the initial resin viscosity, is the resin viscosity at the current moment, is the resin viscosity influence constant;

[0086] where the curing degree of the finished product at the current moment Specifically, it is the curing degree monitored in real time through techniques such as infrared spectroscopy or thermogravimetric analysis. Specifically, Fourier transform infrared spectroscopy (FTIR) is a technique for analyzing materials using infrared radiation. During the curing process, by monitoring the absorption of infrared light of specific wavelengths by the material, information about chemical reactions can be obtained. Curing reactions usually involve changes in certain functional groups, such as the formation or breakage of hydrogen bonds, the conversion of double bonds, etc. By tracking the absorption peaks of these functional groups before and after curing, the curing degree can be inferred. However, due to the influence of environmental factors such as temperature and light conditions during the infrared spectrum data collection process, the disappearance or transformation of some functional groups may not be fully reflected in the infrared spectrum during the curing process. For example, some reactions may result in the formation of new chemical structures or changes in functional groups, and the infrared spectrum may not fully capture these changes, especially in the case of weak absorption at low concentrations or specific wavelengths. As a result, the curing degree detected by the infrared spectrum data leads to a smaller curing degree of the finished product. Therefore, there is an error in determining the curing degree of the finished product only through the infrared spectrum. By combining the change of resin viscosity gradually increasing with the curing reaction and determining the curing degree of the finished product through the infrared spectrum for comprehensive judgment, the error can be compensated and the accuracy of curing degree detection can be improved.

[0087] It should be noted that represents the accurate value of the curing degree of the finished product at the current moment. The curing degree is usually used to describe the degree of transformation of a material from a liquid or semi-solid state to a solid state during the curing process The larger the value, the higher the degree of curing of the material, and the more fully the chemical reactions (such as cross-linking, polymerization, etc.) within the material have proceeded

[0088] Among them, the viscosity of the resin changes during the curing process as the reaction progresses. Generally, as curing proceeds, the viscosity increases. This change in viscosity directly affects the fluidity of the resin and the efficiency of the curing reaction is used to compare the relationship between the viscosity at the current moment and the initial viscosity. The square root of this ratio can be regarded as a proportional correction to the curing degree, reflecting the influence of the current resin state on the curing degree. The square root form is used to adjust the curing degree, which can more smoothly reflect the influence of viscosity changes. The change in viscosity is usually not linear, and using the square root can, to a certain extent, reduce the influence on the curing degree, especially when the change in viscosity is not significant

[0089] The resin viscosity influence constant is used to adjust the calculation of the curing degree to reflect the influence under specific process flows or conditions, and generally takes a value between 0.02 and 0.3

[0090] According to the accurate value of the curing degree of the finished product, the predicted value of the thermal conductivity, and the process flow parameters, the heat demand coefficient, the resin reaction degree coefficient, and the pressure loss coefficient are calculated. The formula based on which the heat demand coefficient is calculated is as follows

[0091]

[0092] In the formula is the heat demand coefficient at the current moment is the air flow rate of the processing environment at the current moment is the cumulative reaction heat at the current moment is the predicted value of the thermal conductivity of the finished product

[0093] It should be noted that the heat demand coefficient at the current moment is characterized by the air flow rate, the cumulative reaction heat, and the predicted value of the thermal conductivity of the finished product. The heat demand coefficient at the current moment The larger the value, the higher the heat required for the reaction, and the mold temperature can be increased to meet the requirements of the curing reaction

[0094] The air flow rate of the processing environment at the current moment , the air flow rate has a significant impact on heat conduction and convective heat transfer. The flowing air can carry away or transfer heat. Air flow will carry away excess heat or introduce cold air, thus affecting the temperature distribution and reaction rate of the material. Therefore, the heat demand coefficient at the current moment is directly proportional to the air flow rate of the processing environment at the current moment and is represented by an exponential function indicating that the increase in air flow rate has a non-linear impact on heat consumption. A higher air flow rate will significantly increase heat loss.

[0095] The cumulative reaction heat at the current moment is larger, indicating that more heat is generated. Therefore, the mold temperature can be reduced to maintain a constant temperature while reducing energy consumption. Thus, the cumulative reaction heat at the current moment is inversely proportional to the heat demand coefficient at the current moment and is represented by a square indicating the inhibitory effect of reaction heat on temperature increase. As the reaction heat increases, the material temperature rises, and the mold temperature can be reduced to reduce consumption.

[0096] The predicted value of the finished product's thermal conductivity , a parameter describing the heat conduction ability of the material. Materials with high thermal conductivity can transfer heat more effectively. Therefore, only a relatively low temperature supply is required to achieve the target effect. Thus, the cumulative reaction heat at the current moment is inversely proportional to the heat demand coefficient at the current moment Taking the logarithm of the thermal conductivity can smooth the impact of thermal conductivity on heat transfer efficiency. The higher the thermal conductivity, the more effectively the material can transfer heat, resulting in an increase in heat transfer efficiency.

[0097] Through it represents the combined effect of cumulative reaction heat and thermal conductivity on temperature control. The square root is used to smooth the impact of the combined effect.

[0098] where the cumulative reaction heat at the current moment is calculated through the heat flux density. The specific formula for the calculation is:

[0099]

[0100] In the formula, is the initial reaction heat, is the start time of processing, is the current moment, is the heat flux density, where the heat flux density is collected in real time through a heat flow meter. The initial reaction heat is generally 0.

[0101] The formula for calculating the resin reaction degree coefficient is:

[0102]

[0103] In the formula, is the resin reaction degree coefficient at the current moment, is the environmental humidity;

[0104] It should be noted that the resin reaction degree coefficient at the current moment, the larger the value, the higher the resin reaction degree and the more complete the curing reaction.

[0105] Among them the larger the value, the higher the curing degree of the material, and the more sufficient the chemical reactions (such as cross-linking, polymerization, etc.) in the material have proceeded. Therefore, the resin reaction degree coefficient is proportional to indicating the progress of the reaction degree.

[0106] The viscosity of the resin changes during the curing process as the reaction proceeds. Generally, as curing progresses, the viscosity increases. Therefore, the higher the resin viscosity the higher the reaction progress. Thus, the resin viscosity is proportional to the resin reaction degree coefficient at the current moment.

[0107] By combining the degree of cure and viscosity, it characterizes the reaction degree of the resin. The degree of cure provides information about the progress of the reaction, while the resin viscosity reflects the change in the state of the resin during the curing process. Using the square root form can balance the influence of these two parameters and avoid one parameter being numerically too dominant. Through such a combination, it can more comprehensively reflect the reaction degree of the resin.

[0108] Humidity will affect the curing process of the resin. Especially in some specific types of resins (such as waterborne resins), their reaction characteristics may be significantly affected by the environmental humidity, resulting in a slowdown in the curing progress. Therefore, the environmental humidity is inversely proportional to the resin reaction degree coefficient at the current moment. Through the logarithmic function it reflects the inhibitory effect of humidity on the resin curing process. Excessive humidity may cause the resin curing reaction rate to slow down or the curing to be incomplete. The introduction of the logarithmic function can smooth the influence of humidity, especially under low humidity conditions, where the influence of humidity changes on the curing process will be more significant.

[0109] The formula based on which the pressure loss coefficient is calculated is:

[0110]

[0111] In the formula, is the pressure loss coefficient of the injection resin at the current moment, is the length of the resin injection channel, is the diameter of the resin injection channel, is the density of the resin, is the resin injection speed of the resin, is the friction factor.

[0112] During the flow process, the resin will be subject to friction and local flow resistance, resulting in pressure loss. The Darcy - Weisbach formula is used to describe the pressure loss. The pressure loss coefficient of the injection resin at the current moment. The larger the value, the more pressure loss in the processing. Therefore, the pressure should be increased to prevent the phenomena of bubbles and gas accumulation. At the same time, the pressure helps to reduce the voids in the material and improve the density and mechanical properties of the final product.

[0113] Among them, the friction factor can be set according to the material of the injection channel combined with expert experience, generally between 0.01 and 0.1.

[0114] Step 5: Dynamically correct the process control parameters in the processing according to the obtained heat demand coefficient, resin reaction degree coefficient and pressure loss coefficient to obtain the real - time correction value of the process control parameters. Control the processing based on the obtained real - time correction value of the process control parameters to complete the process control of composite material forming. The process control parameters include mold temperature, curing pressure and curing time.

[0115] Dynamically correct the process control parameters in the processing according to the obtained heat demand coefficient, resin reaction degree coefficient and pressure loss coefficient to obtain the real - time correction value of the process control parameters. The specific formula for calculating the correction value of the mold temperature is as follows:

[0116]

[0117] In the formula, is the correction value of the mold temperature at the current moment, is the initial value of the set mold temperature, is the heat demand coefficient at the current moment;

[0118] Among them, the heat demand coefficient at the current moment. The larger the value, the higher the heat demand. The mold temperature can be increased to meet the needs of the curing reaction. Therefore, the correction value of the mold temperature at the current moment is proportional to expressed by the logarithmic function With the increase of the influence on the die temperature correction value gradually decreases.

[0119] The resin reaction degree coefficient at the current moment The larger the value, the higher the resin reaction degree, the more complete the curing reaction. On the contrary, the curing reaction is weaker. Therefore, the temperature should be increased to promote the curing reaction. Thus is inversely proportional to and is represented by the inverse relationship, and the square root is used to avoid excessive correction.

[0120] The specific formula for calculating the curing time correction value is:

[0121]

[0122] In the formula, is the curing time correction value at the current moment, is the initial value of the set curing time;

[0123] The resin reaction degree coefficient at the current moment The larger the value, the higher the resin reaction degree, the more complete the curing reaction. The curing time can be appropriately reduced. On the contrary, the curing time should be increased to make the curing reaction complete. Therefore, the curing time correction value at the current moment is inversely proportional to and is represented by the square to indicate the significant influence on the curing time correction value.

[0124] The specific formula for calculating the curing pressure correction value is:

[0125]

[0126] In the formula, is the curing pressure correction value at the current moment, is the initial value of the set curing pressure, where , and are the weight constants of the heat demand coefficient, resin reaction degree coefficient and pressure loss coefficient respectively, where and , and are all greater than 0.

[0127] Among them, the pressure loss coefficient of the injected resin at the current moment The larger the value, the more pressure loss in the processing. Therefore, the pressure should be increased. Thus, the curing pressure correction value at the current moment is proportional to and is represented by the exponential function Indicates a significant impact on pressure control. While the temperature increases under constant volume, the pressure also increases. Therefore, when the temperature rises, the pressure should be appropriately reduced to maintain a stable pressure. Thus, the curing pressure correction value at the current moment is inversely proportional to

[0128] Among them, the heat transfer efficiency usually has the greatest impact on the entire curing process, so it is given the highest weight. The heat transfer efficiency directly affects the temperature and reaction rate of the material, and has a significant decisive effect on the mold temperature and curing pressure. The reaction degree of the resin also has an important impact on the curing process, affecting the curing time and temperature, but its influence usually comes after heat transfer. The adjustment of the curing pressure is often determined by flow characteristics and hydrodynamic factors. In contrast, it has a relatively small direct impact on the mold temperature and reaction dynamics. Therefore, set and 、 and are all greater than 0.

[0129] Please refer to Figure 2 , the present invention also provides a process control system for composite material molding. The process control system for composite material molding is used to execute the above-mentioned process control method for composite material molding, including:

[0130] A sample data acquisition module, which is used to obtain composite material finished products produced with several different types and different addition amounts of fillers, detect the composite material finished products through experiments, obtain the corresponding thermal conductivity of the finished products, map the type and addition amount of the fillers to the corresponding thermal conductivity of the finished products one by one, and generate a training sample data set;

[0131] A prediction model training module, which is used to establish a neural network prediction model based on the data in the training sample data set, use the type and addition amount of the fillers in the training sample data set as the input of the model, and use the corresponding thermal conductivity of the finished products as the label to train the neural network prediction model to obtain a thermal conductivity prediction model;

[0132] A process parameter analysis module, which is used to obtain the type and addition amount of the fillers added to the composite material finished product to be produced, input the type and addition amount of the fillers added to the composite material finished product to be produced into the trained thermal conductivity prediction model to obtain the predicted value of the thermal conductivity of the finished product to be produced, and simultaneously obtain the real-time process flow parameters during the processing. The process flow parameters include resin viscosity, heat flux density, environmental humidity, resin injection speed, and air flow rate;

[0133] The curing degree detection module is used to obtain the real-time curing degree of the finished product during the processing, correct the curing degree of the finished product based on the process flow parameters to obtain the accurate value of the curing degree of the finished product, and calculate the heat demand coefficient, resin reaction degree coefficient and pressure loss coefficient according to the accurate value of the curing degree of the finished product, the predicted value of the thermal conductivity and the process flow parameters;

[0134] The control parameter correction module is used to dynamically correct the process control parameters during the processing according to the obtained heat demand coefficient, resin reaction degree coefficient and pressure loss coefficient to obtain the real-time correction value of the process control parameters, and control the processing based on the obtained real-time correction value of the process control parameters to complete the process control of composite material forming. The process control parameters include mold temperature, curing pressure and curing time.

[0135] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the real situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0136] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in this article can be implemented by electronic hardware, or by the combination of computer software and electronic hardware. Whether these functions are executed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0137] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units. They can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0138] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application, and all of them should be covered within the protection scope of this application.

Claims

1. A process control method for forming a composite material, characterized in that, The specific steps include: Obtain finished composite materials produced with several different types and addition amounts of fillers, detect the finished composite materials through experiments, obtain the corresponding thermal conductivity of the finished products, map the types and addition amounts of the fillers one by one to the corresponding thermal conductivity of the finished products, and generate a training sample data set; Based on the data in the training sample data set, establish a neural network prediction model. Use the types and addition amounts of the fillers in the training sample data set as the input of the model, and use the corresponding thermal conductivity of the finished products as the label to train the neural network prediction model to obtain a thermal conductivity prediction model; Obtain the types and addition amounts of the fillers added to the finished composite materials to be produced, input the types and addition amounts of the fillers added to the finished composite materials to be produced into the trained thermal conductivity prediction model to obtain the predicted value of the thermal conductivity of the finished products to be produced. At the same time, obtain the real-time process flow parameters during the processing, and the process flow parameters include resin viscosity, heat flux density, environmental humidity, resin injection speed, and air flow rate; Obtain the real-time degree of cure of the finished products during the processing, correct the degree of cure of the finished products based on the process flow parameters to obtain the accurate value of the degree of cure of the finished products. Calculate the heat demand coefficient, resin reaction degree coefficient, and pressure loss coefficient according to the accurate value of the degree of cure of the finished products, the predicted value of the thermal conductivity, and the process flow parameters; Obtain the real-time degree of cure of the finished products during the processing, correct the degree of cure of the finished products based on the process flow parameters to obtain the accurate value of the degree of cure of the finished products. The formula for calculating the accurate value of the degree of cure of the finished products is: Wherein, is the accurate value of the curing degree of the finished product at the current moment, is the curing degree of the finished product collected at the current moment, is the initial resin viscosity, is the resin viscosity at the current moment, is the resin viscosity influence constant; wherein the curing degree of the finished product at the current moment is monitored in real time by infrared spectroscopy or thermogravimetric analysis technology; Dynamically correct the process control parameters during the processing according to the obtained heat demand coefficient, resin reaction degree coefficient, and pressure loss coefficient to obtain the real-time correction value of the process control parameters. Control the processing based on the obtained real-time correction value of the process control parameters to complete the process control of composite material forming. The process control parameters include mold temperature, curing pressure, and curing time; 2. The process control method for forming a composite material according to claim 1, characterized in that: Among them, the generation method of the training sample data set is: map the types and addition amounts of the fillers one by one to the thermal conductivity parameters of the corresponding finished composite materials to form a corresponding grid, and record the formed grid as the training sample data set; Based on a convolutional neural network, a thermal conductivity prediction model is established. The convolutional neural network consists of an input layer, a convolutional layer, a pooling layer, a fully connected layer, and an output layer. The activation function in the convolutional layer is function, The specific expression of the function is: Among them, represents the th corresponding convolutional layer, represents the th eigenvalue of the th feature data in the th corresponding convolutional layer, where is the index of the convolutional layer, is the index of the feature data, is the index of the eigenvalue in the feature data, where , , , where N, M, and O are the total numbers of convolutional layers, feature data, and eigenvalues in the feature data, respectively; For the fully connected layer, set the number of neurons in the fully connected layer to 32, set the initial neural network learning rate to 0.001, and the number of training epochs to 100; The input of the trained thermal conductivity prediction model is the type and addition amount of the filler, and the output is the predicted value of the thermal conductivity of the finished composite material; 3. A process control method for forming a composite material according to claim 1, characterized in that: Calculate the heat demand coefficient, resin reaction degree coefficient, and pressure loss coefficient according to the accurate value of the degree of cure of the finished products, the predicted value of the thermal conductivity, and the process flow parameters. The formula for calculating the heat demand coefficient is: In the formula, is the heat demand coefficient at the current moment, is the air flow rate of the processing environment at the current moment, is the cumulative reaction heat at the current moment, is the predicted value of the thermal conductivity of the finished product; The cumulative heat of reaction at the current moment is calculated through the heat flux density, and the specific formula for the calculation is as follows: In the formula, is the initial reaction heat, is the start time of processing, is the current time, is the heat flux density, and the heat flux density is collected in real time by a heat flux meter.

4. A process control method for forming a composite material according to claim 2, characterized in that: Among them, the formula for calculating the resin reaction degree coefficient is: In the formula, is the resin reaction degree coefficient at the current moment, is the environmental humidity; Among them, the formula for calculating the pressure loss coefficient is: Wherein, is the pressure loss coefficient of the injection resin at the current moment, is the length of the resin injection runner, is the diameter of the resin injection runner, is the density of the resin, is the resin injection speed of the resin, is the friction factor.

5. A process control method for forming a composite material according to claim 4, characterized in that: Dynamically correct the process control parameters during the processing according to the obtained heat demand coefficient, resin reaction degree coefficient, and pressure loss coefficient to obtain the real-time correction value of the process control parameters. The formula specifically used for calculating the correction value of the mold temperature is: In the formula, is the mold temperature correction value at the current moment, is the initial value of the set mold temperature, is the heat demand coefficient at the current moment; Among them, the formula specifically used for calculating the correction value of the curing time is: Wherein, is the curing time correction value at the current moment, is the initial value of the set curing time; The specific formula for calculating the curing pressure correction value is as follows: In the formula, is the curing pressure correction value at the current moment, is the initial curing pressure value set, where , and are the weight constants of the heat demand coefficient, the resin reaction degree coefficient, and the pressure loss coefficient respectively, where and , and are all greater than 0.

6. A process control system for forming a composite material, characterized in that: The process control system for composite material forming is used to execute the process control method for composite material forming according to any one of claims 1-5, and includes: A sample data acquisition module, which is used to obtain composite material finished products produced with several different types and addition amounts of fillers, detect the composite material finished products through experiments, obtain the corresponding thermal conductivity of the finished products, map the type and addition amount of the fillers to the corresponding thermal conductivity of the finished products one by one, and generate a training sample data set; A prediction model training module, which is used to establish a neural network prediction model based on the data in the training sample data set, use the type and addition amount of the fillers in the training sample data set as the input of the model, and use the corresponding thermal conductivity of the finished products as the label to train the neural network prediction model to obtain a thermal conductivity prediction model; A process parameter analysis module, which is used to obtain the type and addition amount of the fillers added to the composite material finished product to be produced, input the type and addition amount of the fillers added to the composite material finished product to be produced into the trained thermal conductivity prediction model to obtain the predicted value of the thermal conductivity of the finished product to be produced, and at the same time obtain the real-time process flow parameters during the processing, and the process flow parameters include resin viscosity, heat flux density, environmental humidity, resin injection speed and air flow rate; A curing degree detection module, which is used to obtain the real-time curing degree of the finished product during the processing, correct the curing degree of the finished product based on the process flow parameters to obtain the accurate value of the curing degree of the finished product, and calculate the heat demand coefficient, resin reaction degree coefficient and pressure loss coefficient according to the accurate value of the curing degree of the finished product, the predicted value of the thermal conductivity and the process flow parameters; A control parameter correction module, which is used to dynamically correct the process control parameters during the processing according to the obtained heat demand coefficient, resin reaction degree coefficient and pressure loss coefficient to obtain the real-time correction value of the process control parameters, and control the processing based on the obtained real-time correction value of the process control parameters to complete the process control of composite material forming, and the process control parameters include mold temperature, curing pressure and curing time.

Images