Impregnation amount control method for alkali-resistant glass fiber gridding cloth

Abstract

The invention provides an alkali-resistant fiberglass mesh impregnation amount control method which comprises the following steps: acquiring surface topography structure parameters and chemical property parameters of fiberglass mesh, judging wettability and affinity of a fiber surface according to a preset impregnation process parameter threshold, and determining an initial impregnation amount range; plasma treatment is carried out on the surface of the glass fiber gridding cloth, hydrophilic groups are introduced, the fiber surface energy is improved, the affinity with a glue solution is enhanced, and high glue dipping sufficiency is achieved under the condition of low glue dipping amount; according to structural parameters such as the fiber diameter and the fiber spacing of the fiberglass mesh, a finite element analysis method is adopted, the spreading state of a glue solution on the fiber surface and the penetration depth in a fiber bundle under different glue dipping amounts are simulated, and the quantitative relation between the glue dipping amount and the glue dipping sufficiency is obtained.

Description

Technical Field

[0001] The invention relates to the field of information technology, and in particular to a method for controlling the amount of glue impregnation of an alkali-resistant glass fiber mesh cloth. Background Art

[0002] Background: During the dipping process of glass fiber mesh cloth, the wetting and spreading of the glue on the fiber surface and the penetration of the fiber bundle are the key factors affecting the dipping amount. If the dipping amount is too low, the bonding force between the fiber and the matrix interface will be insufficient, and it will be difficult to fully exert the reinforcing effect; if the dipping amount is too high, the glue will easily form a thick coating on the fiber surface, causing the mesh cloth to harden and become brittle, and the alkali resistance will decrease.

[0003] During the dipping process, process parameters such as glue viscosity, surface tension, dipping time, and dipping times will affect the dipping amount. When the glue viscosity is low, it is easy to penetrate the fiber bundle, but it is easy to drip and it is difficult to control the dipping amount; when the glue surface tension is low, it has good wetting and spreading properties, but poor penetration. When the dipping time is long, the glue penetrates fully, but it is easy to cause local excessive dipping; when the dipping times are many, the glue is evenly distributed, but the cost is high and the efficiency is low.

[0004] In addition, the chemical properties and morphological structure of the glass fiber mesh surface also have a significant impact on the amount of impregnation. The fiber surface energy is low, the affinity with the glue is poor, and the amount of impregnation is low; the fiber surface is rough and the specific surface area is large, which is conducive to the adsorption of the glue, but the gaps between the fibers are small, which is not conducive to the penetration of the glue. Therefore, how to optimize the impregnation process and fiber surface modification, find a balance between the adequacy of impregnation and the uniformity of glue distribution, and control the impregnation amount within the optimal range is a key technical problem that needs to be solved in the control of the impregnation amount of glass fiber mesh. Summary of the invention

[0005] The present invention provides a method for controlling the amount of glue impregnation of an alkali-resistant glass fiber mesh cloth, which mainly comprises: The surface morphology structural parameters and chemical property parameters of the glass fiber mesh cloth are obtained, and the wettability and affinity of the fiber surface are judged according to the preset threshold value of the dipping process parameters, and the initial dipping amount range is determined; plasma treatment is performed on the surface of the glass fiber mesh cloth to introduce hydrophilic groups, increase the fiber surface energy, enhance the affinity with the glue solution, and achieve high dipping sufficiency under low dipping amount; according to the structural parameters of the glass fiber mesh cloth such as fiber diameter and fiber spacing, the finite element analysis method is used to simulate the spreading state of the glue solution on the fiber surface and the penetration depth inside the fiber bundle under different dipping amounts, and the quantitative relationship between the dipping amount and the dipping sufficiency is obtained; based on the above quantitative relationship between the dipping amount and the dipping sufficiency, by changing the type and amount of additives in the glue solution formula, the viscosity and surface tension of the glue solution are adjusted, and the change law of the uniformity of the glue solution distribution under different dipping amounts is obtained, and the optimal glue solution formula is determined in combination with the requirements of dipping sufficiency, and the uniformity of the glue solution distribution is quantitatively measured by optical microscope observation and image analysis software. The coverage rate and thickness distribution of the adhesive on the fiber surface are used to describe it; for the determined optimal adhesive formula and different dipping amounts, the orthogonal experimental design method is used to optimize the process parameters such as dipping time and number of dipping times, and the best matching scheme between the dipping amount and the process parameters is screened out through tensile strength, alkali resistance, bending strength, and shear strength performance tests; a nonlinear prediction model between the dipping amount, adhesive performance, fiber surface properties, dipping process parameters and dipping effect is established by combining numerical simulation with machine learning. The optimal combination of various parameters is found through the optimization algorithm to guide the precise control of the dipping amount. The dipping effect includes the adequacy of dipping, uniformity of adhesive distribution, internal penetration depth of the fiber bundle and the mechanical properties of the final composite material; in actual production, the dipping amount and process parameters are set according to the prediction model, and the real-time feedback control of the dipping amount is achieved through online monitoring of parameters such as adhesive viscosity and surface tension, combined with the detection of the surface morphology of the mesh cloth after dipping by machine vision, to ensure the stability of the dipping quality.

[0006] The technical solution provided by the embodiment of the present invention may have the following beneficial effects: The present invention discloses a high-performance dipping method for glass fiber mesh cloth with low dipping amount. The method first obtains the surface structure and chemical property parameters of the mesh cloth, judges the fiber wettability in combination with a preset process threshold, and determines the initial dipping amount range. Then, the fiber surface energy is increased by plasma treatment to enhance the affinity with the glue solution. Finite element analysis is used to simulate the distribution state of the glue solution under different dipping amounts, and a quantitative relationship between the dipping amount and the sufficiency is established. Based on this relationship, the optimal distribution uniformity is obtained by adjusting the glue solution formula. Orthogonal experiments are further used to optimize the process parameters and screen out the best matching scheme. Finally, a nonlinear prediction model is established to guide the precise control of the dipping amount and realize real-time feedback regulation in production. The present invention achieves high dipping sufficiency and uniformity at low dipping amounts, improves the mechanical properties of composite materials, and has important engineering application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 The present invention is a flow chart of a method for controlling the amount of glue impregnation of an alkali-resistant glass fiber mesh cloth.

[0008] Figure 2 The present invention is a schematic diagram of a method for controlling the amount of glue impregnation of an alkali-resistant glass fiber mesh cloth.

[0009] Figure 3 It is another schematic diagram of a method for controlling the amount of glue impregnation of an alkali-resistant glass fiber mesh cloth according to the present invention. DETAILED DESCRIPTION

[0010] In order to further understand the content of the present invention, the present invention is described in detail in conjunction with the accompanying drawings and embodiments. The present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are only used to explain the relevant inventions, rather than to limit the invention. It is also necessary to explain that, for ease of description, only the parts related to the invention are shown in the accompanying drawings.

[0011] like Figure 1-3 In this embodiment, a method for controlling the amount of glue impregnation of an alkali-resistant glass fiber mesh cloth may specifically include: Step S101, obtaining the surface morphology structure parameters and chemical property parameters of the glass fiber mesh cloth, judging the wettability and affinity of the fiber surface according to the preset dipping process parameter threshold, and determining the initial dipping amount range.

[0012] The surface morphology image of the glass fiber mesh cloth is obtained by scanning electron microscopy, and the structural parameters of the fiber surface, including fiber diameter, porosity, etc., are extracted by image processing algorithm. The chemical property parameters of the glass fiber surface, including surface element composition, functional group type, etc., are tested by X-ray photoelectron spectrometer. According to the preset threshold of the dipping process parameters, it is judged whether the fiber surface structural parameters and chemical property parameters meet the requirements. If they meet the requirements, it is determined that the fiber surface has good wettability and resin affinity. According to the wettability and affinity of the fiber surface, combined with the dipping process parameters, the dipping amount prediction model is established by support vector machine regression algorithm to obtain the reasonable range of the initial dipping amount. If the actual dipping amount is lower than the lower limit of the prediction range, the dipping amount is increased by increasing the extrusion force of the dipping roller; if the actual dipping amount is higher than the upper limit of the prediction range, the dipping amount is reduced by reducing the extrusion force of the dipping roller. During the dipping process, the dipping amount is monitored in real time by machine vision, and the monitoring data is input into the dipping amount prediction model for comparison, and the dipping process parameters are dynamically adjusted to ensure that the dipping amount is always within a reasonable range. Genetic algorithm is used to optimize the dipping process parameters, with the stability of dipping amount and glass fiber performance as the optimization goals, to obtain the best combination of dipping process parameters to guide production practice.

[0013] Exemplarily, the surface morphology image of the glass fiber mesh cloth is obtained by scanning electron microscopy, and the structural parameters of the fiber surface are extracted by using the Canny edge detection algorithm, including the fiber diameter between 5-10μm and the porosity between 20%-30%. The chemical property parameters of the glass fiber surface are tested by X-ray photoelectron spectrometer, including the surface carbon content of 60%-70%, the oxygen content of 20%-30%, the silicon content of 10%-20%, and the surface hydroxyl content of 5%-10%. According to the preset dip process parameter threshold, it is judged that the fiber diameter, porosity, surface element composition and functional group content all meet the requirements, and it is determined that the fiber surface has good wettability and epoxy resin affinity. According to the wettability and affinity of the fiber surface, combined with the process parameters such as the dip temperature of 60℃-80℃ and the dip time of 10-20 minutes, the dip amount prediction model is established by the support vector machine regression algorithm, and the reasonable range of the initial dip amount is 150-200g / m². If the actual amount of dipping glue is lower than 150g / m², the dipping glue amount is increased by increasing the extrusion pressure of the dipping glue roller to 5-0MPa; if the actual amount of dipping glue is higher than 200g / m², the dipping glue amount is reduced by reducing the extrusion pressure of the dipping glue roller to 1-3MPa. During the dipping process, the dipping glue amount is monitored in real time by machine vision, and the monitoring data is input into the dipping glue amount prediction model for comparison. If the deviation exceeds ±10%, the extrusion pressure of the dipping glue roller is dynamically adjusted to ensure that the dipping glue amount is always within the range of 150-200g / m². The genetic algorithm is used to optimize the dipping glue process parameters, with the stability of the dipping glue amount and the tensile strength of the glass fiber as the optimization goals. By constructing the fitness function, the crossover probability is selected as 8, the mutation probability is 1, and the evolution is carried out for 100 generations. The best combination of dipping glue process parameters is obtained, which is a dipping glue temperature of 75℃, a dipping glue time of 15 minutes, and a dipping glue roller extrusion pressure of 6MPa. It guides production practice and improves the stability of the dipping glue amount by 20% and the tensile strength of the glass fiber by 15%.

[0014] Step S102, plasma treatment is performed on the surface of the glass fiber mesh cloth to introduce hydrophilic groups, increase the surface energy of the fiber, enhance the affinity with the glue solution, and achieve high glue sufficiency at a low glue amount.

[0015] Obtain a glass fiber mesh cloth, place it in a plasma treatment device for surface treatment, introduce hydrophilic groups to the fiber surface through plasma, and increase the fiber surface energy. According to the material and structural characteristics of the glass fiber mesh cloth, determine the appropriate plasma treatment process parameters, including power, pressure, gas components, etc., to introduce sufficient hydrophilic groups on the fiber surface while avoiding damage to the fiber. Perform a surface energy test on the glass fiber mesh cloth after plasma treatment to determine whether it meets the preset hydrophilicity requirements. If not, return to step 2 to adjust the process parameters and reprocess. Immerse the hydrophilized glass fiber mesh cloth in the glue solution, and through the affinity between the fiber surface and the glue solution, the glue solution is fully infiltrated and adhered to the fiber surface under low dipping amount. Use machine vision technology to detect the glass fiber mesh cloth after dipping, obtain the distribution of the glue solution on the fiber surface, and determine whether the uniformity and sufficiency of the dipping meet the requirements. If the quality of the dipping glue does not meet the standard, the affinity between the glue and the hydrophilized fiber surface can be improved by optimizing the glue formula and adjusting the dipping process parameters, such as dipping time and glue viscosity, to improve the dipping effect. The glass fiber mesh cloth with qualified dipping glue is sent to the drying equipment and dried under controlled temperature and time conditions to solidify the glue and firmly adhere to the fiber surface, so as to obtain a reinforced glass fiber mesh cloth product with high dipping adequacy.

[0016] Exemplarily, the glass fiber mesh is first arranged in a plasma treatment device, and a mixed gas of oxygen and argon is used to perform surface treatment under the conditions of a power of 500W and a pressure of 50Pa. Hydrophilic groups such as hydroxyl and carboxyl are introduced by plasma to increase the fiber surface energy from 32mJ / m² to 58mJ / m². The surface energy of the treated glass fiber mesh cloth is then tested by the contact angle method, and the surface energy is calculated by the Young-Dupre equation. If the target value of 58mJ / m² is not reached, the plasma treatment process parameters are optimized by orthogonal experiments, and infrared spectroscopy is used to analyze the changes in chemical groups on the fiber surface until the requirements are met. The hydrophilized glass fiber mesh cloth is immersed in a glue solution composed of epoxy resin, curing agent and additives. Through capillary force and hydrogen bonding, the glue solution is fully infiltrated and adhered to the fiber surface when the dipping amount is less than 5%. The machine vision system was used to collect images of the glass fiber mesh cloth surface after dipping. The glue distribution area was extracted by image segmentation algorithm, and the glue coverage was calculated. If it was less than 95%, it was judged as uneven dipping. The glue formula was optimized by response surface method, the glue viscosity was adjusted to 500mPa·s, and the dipping time was controlled to 60s to improve the dipping effect. Finally, the glass fiber mesh cloth with qualified dipping was sent to the drying equipment and dried at 80℃ for 30min to solidify the glue and firmly adhere to the fiber surface. After testing, the tensile strength of the glass fiber mesh cloth increased by 30%, achieving the enhancement effect of high dipping sufficiency.

[0017] Step S103, based on the fiber diameter, fiber spacing and other structural parameters of the glass fiber mesh cloth, a finite element analysis method is used to simulate the spreading state of the glue on the fiber surface and the penetration depth inside the fiber bundle under different glue impregnation amounts, so as to obtain a quantitative relationship between the glue impregnation amount and the glue impregnation sufficiency.

[0018] According to the structural parameters of the glass fiber mesh cloth, such as fiber diameter and fiber spacing, a finite element analysis model is established. Different impregnation amounts are set to simulate the spreading state of the glue on the fiber surface. If the glue does not completely cover the fiber surface, the impregnation amount is increased and the spreading state of the glue is re-simulated until the glue completely covers the fiber surface. According to the simulation results, the spreading area of ​​the glue on the fiber surface under different impregnation amounts is obtained. The penetration depth of the glue inside the fiber bundle under different impregnation amounts is simulated. If the penetration depth does not reach the preset threshold, the impregnation amount is increased and the penetration depth is re-simulated until the preset threshold is reached. Through multiple simulations, the spreading area of ​​the glue on the fiber surface and the penetration depth inside the fiber bundle under different impregnation amounts are obtained, and a quantitative relationship model between the impregnation amount and the adequacy of the impregnation is established.

[0019] Exemplarily, first, according to the structural parameters of the glass fiber mesh cloth such as fiber diameter and fiber spacing, a finite element analysis model is established. For example, the fiber diameter is 10 microns, the fiber spacing is 100 microns, and tetrahedral meshing is used, and the mesh size is 5 microns. Then set different dipping amounts, such as 1ml / cm², 2ml / cm², 3ml / cm², etc., and use fluid mechanics software such as FLUENT to simulate the spreading state of the glue on the fiber surface. By comparing the area ratio of the glue covering the fiber surface under different dipping amounts, it is judged whether the glue completely covers the fiber surface. If the coverage area ratio does not reach 100%, the dipping amount is increased, and the glue spreading state is re-simulated until the glue completely covers the fiber surface. According to the simulation results, the spreading area of ​​the glue on the fiber surface under different dipping amounts is obtained, such as 80% at 1ml / cm², 95% at 2ml / cm², and 100% at 3ml / cm². Then, the penetration depth of the glue solution inside the fiber bundle under different amounts of dipping was simulated. Assuming that the diameter of the fiber bundle is 500 microns, the preset penetration depth threshold is 80% of the fiber bundle radius, that is, 200 microns. The capillary penetration model is used to calculate the penetration depth under different amounts of dipping. If the penetration depth does not reach 200 microns, the dipping amount is increased and the penetration depth is re-simulated until the preset threshold is reached. Through multiple simulations, the spreading area of ​​the glue solution on the fiber surface and the penetration depth inside the fiber bundle under different dipping amounts are obtained, such as 1ml / cm² when the spreading area is 80% and the penetration depth is 150 microns; 2ml / cm² when the spreading area is 95% and the penetration depth is 180 microns; 3ml / cm² when the spreading area is 100% and the penetration depth is 210 microns. Finally, based on the above data, the least squares method is used to fit the functional relationship between the dipping amount, the spreading area, and the penetration depth, and a quantitative relationship model between the dipping amount and the dipping sufficiency is established, providing a theoretical basis for process optimization.

[0020] Step S104, based on the quantitative relationship between the amount of dipping and the adequacy of dipping, by changing the type and amount of the additives in the glue solution formula, adjusting the viscosity and surface tension of the glue solution, obtaining the change law of the glue solution distribution uniformity under different dipping amounts, and combining the requirements of dipping adequacy, determining the best glue solution formula. The glue solution distribution uniformity is described by optical microscope observation and quantitative calculation of the glue solution coverage and thickness distribution on the fiber surface by image analysis software.

[0021] According to the preset glue formula, multiple groups of glue samples with different types and amounts of additives are obtained, and the viscosity and surface tension parameters of each sample are measured to obtain the correlation between the additive parameters and the physical properties of the glue. For different dipping amounts, an optical microscope is used to observe the distribution of multiple glue samples on the fiber surface, and the glue coverage and thickness distribution parameters are calculated by image analysis software to establish a quantitative relationship model between the dipping amount and the uniformity of glue distribution. According to the target requirements of dipping adequacy, the qualified range of glue coverage and thickness distribution is determined, which is used as a constraint condition. Combined with the dipping amount and distribution uniformity model, the optimal dipping amount that meets the constraints is solved by the optimization algorithm. According to the optimal dipping amount, combined with the correlation between the additive parameters and the physical properties of the glue, a machine learning algorithm such as a support vector machine or a neural network is used to establish a glue formula optimization model for predicting and selecting the optimal glue formula. For the preferred glue formula, a verification experiment is carried out, and the glue coverage and thickness distribution are calculated by optical microscope observation and image analysis software to determine whether it meets the requirements of dipping adequacy. If the requirements are met, the formula is determined to be the best glue formula; if not, return to step 4, adjust the optimization model or constraint conditions, and re-optimize the formula until the requirements are met. According to the best glue formula, prepare an appropriate amount of glue for the subsequent fiber dipping process to ensure the quality and efficiency of dipping and improve the performance stability of fiber-reinforced composite materials.

[0022] For example, according to the preset glue formula, 10 groups of glue samples with different types of additives (such as dispersants, tackifiers, etc.) and dosages (1%~0%) were obtained, and the viscosity (500~2000 mPa·s) and surface tension (20~50 mN / m) parameters of each sample were measured by rotational viscometer and tensiometer. The correlation between the additive parameters and the physical properties of the glue was obtained by multivariate linear regression analysis, and a quantitative prediction model was established. For different dipping amounts (50~200 g / m2), the distribution of 20 glue samples on the carbon fiber surface was observed by optical microscopy, and the glue coverage (80%~100%) and thickness distribution parameters (coefficient of variation 1~3) were calculated by image analysis software ImageJ, and a quantitative relationship model between the dipping amount and the uniformity of glue distribution was established by quadratic polynomial fitting. According to the target requirements of the adequacy of the dipping, the qualified range of the glue coverage ≥ 95% and the coefficient of variation of the thickness distribution ≤ 2 was determined. It was used as a constraint condition. Combined with the dipping amount and the distribution uniformity model, the optimal dipping amount that meets the constraints was solved by genetic algorithm to be 120 g / m2. According to the optimal dipping amount, combined with the association law between the auxiliary parameters and the physical properties of the glue, the support vector machine algorithm was used to establish the glue formula optimization model. Through grid search and cross-validation, the optimal glue formula (dispersant 5%, tackifier 3%, other components unchanged) was selected. For the preferred glue formula, three repeated verification experiments were carried out. The glue coverage (97% ± 1%) and the coefficient of variation of the thickness distribution (18 ± 02) were calculated by optical microscope observation and ImageJ software. The formula met the requirements of the adequacy of the dipping and was determined to be the optimal glue formula. According to the optimal glue formula, 2 kg of glue was prepared for the subsequent carbon fiber dipping process to improve the interface performance and mechanical property stability of the composite material.

[0023] Step S105, for the determined optimal glue solution formula and different dipping amounts, an orthogonal experimental design method is used to optimize process parameters such as dipping time and dipping times, and the best matching solution between dipping amount and process parameters is screened out through tensile strength, alkali resistance, bending strength, and shear strength performance tests.

[0024] According to the determined optimal glue formula, different dipping levels are set, and the orthogonal test design method is used to determine the dipping time and dipping times at each level. For each orthogonal test scheme, the corresponding dipping samples are prepared, and performance tests such as tensile strength, alkali resistance, bending strength and shear strength are carried out. According to the performance test results, the variance analysis method is used to judge the significance of the influence of each process parameter on the performance index, and the optimal level combination of each parameter is determined. Through the multi-objective optimization algorithm, the various performance indicators are comprehensively considered to obtain the best matching scheme for dipping time and dipping times under different dipping amounts. A verification test is carried out on the best matching scheme to test whether its performance indicators meet the requirements. If they meet the requirements, it is determined as the final process parameter combination. According to the final determined dipping amount and process parameters, the dipping process specifications are established, and the operating points and precautions of each process are specified to ensure the stability of the production process and the consistency of product quality. Online monitoring technology is used to collect key parameters such as temperature, time, dipping amount, etc. in the dipping process in real time, and the process deviation is timely discovered and corrected through statistical process control methods, and the dipping process is continuously optimized.

[0025] Exemplarily, according to the determined optimal glue solution formula, the dipping amount level is set to 5%, 10%, 15% and 20%, and the L9 (34) orthogonal experimental design method is used to determine that the dipping time at each level is 10min, 20min, 30min, and the dipping times are 1 time, 2 times, and 3 times, respectively. For each orthogonal experimental scheme, a corresponding dipping sample is prepared, and performance tests such as tensile strength, alkali resistance, bending strength and shear strength are carried out. According to the performance test results, the variance analysis method is used to judge the significance of the influence of each process parameter on the performance index. The results show that the dipping amount and dipping time have a significant effect on tensile strength and bending strength, and the dipping times have a significant effect on alkali resistance. It is determined that the optimal level of dipping amount is 15%, the optimal level of dipping time is 20min, and the optimal level of dipping times is 2 times. Through the multi-objective optimization algorithm, the various performance indicators are comprehensively considered, the objective function is established, and the genetic algorithm is used to solve it, and the best matching scheme of dipping time and dipping number under different dipping amounts is obtained, that is, when the dipping amount is 15%, the dipping time is 20min, and the dipping number is 2 times. The best matching scheme is verified by testing that its performance indicators meet the requirements and are determined as the final process parameter combination. According to the final dipping amount and process parameters, the dipping process specification is established, and the operating points and precautions of each process are specified, such as the need to defoam the glue solution before dipping, the need to control the dipping speed to 10mm / min during dipping, and the need to dry at 80℃ for 30min after dipping, etc., to ensure the stability of the production process and the consistency of product quality. The online monitoring technology is used to collect key parameters such as temperature, time, dipping amount, etc. in the dipping process in real time, and the process capability index is calculated by the statistical process control method. When the process capability index is lower than 33, the process deviation is discovered and corrected in time, the dipping process is continuously optimized, and the product quality and production efficiency are continuously improved.

[0026] Step S106, using a method combining numerical simulation and machine learning, a nonlinear prediction model between the amount of glue, glue performance, fiber surface properties, glue process parameters and the glue effect is established, and the optimal combination of each parameter is found through an optimization algorithm to guide the precise control of the amount of glue. The glue effect includes the adequacy of glue, the uniformity of glue distribution, the penetration depth inside the fiber bundle and the mechanical properties of the final composite material.

[0027] Obtain the dipping process parameters, glue performance parameters and fiber surface property parameters, and establish the corresponding mathematical model. Use the numerical simulation method to solve the established mathematical model and obtain the physical field distribution of the dipping process. According to the numerical simulation results, extract the characteristic parameters related to the dipping effect, such as dipping sufficiency, glue distribution uniformity, and fiber bundle internal penetration depth. Obtain a certain number of dipping process parameter combinations and corresponding dipping effect test data as the training data set for the machine learning model. Use machine learning algorithms such as support vector machines or neural networks to establish a nonlinear prediction model between dipping process parameters and dipping effects. Use optimization algorithms such as genetic algorithms, take the dipping effect as the optimization target, find the optimal combination of dipping process parameters, and achieve precise control of the dipping amount. According to the optimized process parameter combination, optimize the dipping process, prepare composite material samples and test their mechanical properties to verify the effectiveness of dipping amount control.

[0028] For example, the dipping process parameters such as glue viscosity 500 mPa·s, dipping speed 10 mm / s, etc., glue performance parameters such as surface tension 35 mN / m, density 2 g / cm³, etc., fiber surface property parameters such as contact angle 60°, surface roughness 5 μm, etc. are obtained, and a mathematical model including Navier-Stokes equation, Darcy's law, etc. is established. The finite element method is used for numerical solution to obtain the distribution of physical fields such as velocity field and pressure field. Characteristic parameters such as dipping sufficiency 95%, glue distribution uniformity coefficient 9, and fiber bundle internal penetration depth 200 μm are extracted. 100 sets of dipping process parameter combinations and corresponding dipping effect test data are obtained, and the support vector machine algorithm is used to establish a prediction model, and the accuracy of 5-fold cross validation reaches 90%. Genetic algorithm optimization is used with a population of 50 and an evolutionary algebra of 100 to obtain the optimal process parameter combination. Composite material samples are prepared, and the tensile strength is increased by 15%, which verifies the effectiveness of dipping amount control.

[0029] Step S107, in actual production, the dipping amount and process parameters are set according to the prediction model, and the viscosity, surface tension and other parameters of the glue solution are monitored online, and the surface morphology of the mesh cloth after dipping is detected by machine vision to achieve real-time feedback control of the dipping amount and ensure the stability of the dipping quality.

[0030] According to the prediction model, the initial dipping process parameters, including dipping time, dipping speed, glue temperature, etc., are set as the reference value for dipping control. Online sensors are used to monitor the performance indicators of the glue solution in real time, including parameters such as glue viscosity and surface tension, to obtain real-time data on the performance of the glue solution. The surface morphology of the mesh cloth after dipping is imaged and analyzed by the machine vision system, and the surface morphology features are extracted to determine whether the dipping amount is uniform. The real-time monitored glue performance data and the surface morphology features analyzed by machine vision are input into the prediction model, and the dipping process parameters are dynamically adjusted to achieve online feedback control of the dipping amount. If the real-time monitored glue performance exceeds the preset threshold range, or the machine vision detects uneven surface morphology, the early warning mechanism is triggered to automatically adjust the dipping process parameters to ensure that the dipping quality meets the requirements. Continuously record and analyze production process data, including the trend of glue performance changes, the distribution of surface morphology features, etc., optimize the prediction model and process parameters, and improve the accuracy and stability of dipping control. Regularly calibrate and maintain the online sensors and machine vision systems to ensure the accuracy and reliability of the monitoring data and provide reliable data support for dipping quality control.

[0031] For example, based on historical production data and quality inspection results, a prediction model between the dipping process parameters and product quality is established, and the support vector machine algorithm is used for training. The dipping time is set to 10 seconds, the dipping speed is 5 meters per second, and the glue temperature is 25°C as the initial process parameters. Online sensors such as viscometers and tensiometers are installed on the production line, and the glue viscosity and surface tension data are collected every 1 second. The data is smoothed by the Kalman filter algorithm to monitor the performance changes of the glue in real time. The surface of the mesh cloth after dipping is scanned by a high-speed industrial camera and line structured light to obtain a three-dimensional surface morphology image, extract characteristic parameters such as surface roughness and height distribution, and determine whether the dipping amount is uniform by the threshold segmentation algorithm. The glue performance data and surface morphology characteristics are input into the prediction model. Based on the model output results, the PID control algorithm is used to dynamically adjust the process parameters such as dipping time, speed and temperature to achieve online feedback control of the dipping amount, and the control accuracy reaches ±1 grams per square meter. Set the warning thresholds for the viscosity and surface tension of the glue solution. When the real-time monitoring data exceeds the threshold range of ±5%, or the uniformity index of the surface morphology is lower than 90%, the warning signal is triggered, and the process parameters are automatically adjusted through the fuzzy control algorithm to ensure that the quality of the dipping glue meets the requirements. Use big data analysis technology to mine and analyze the production process data, determine the key influencing factors between the performance of the glue solution and the surface morphology characteristics and product quality through correlation analysis, and use genetic algorithms to optimize the prediction model and process parameters to improve the accuracy and stability of the dipping glue control. Calibrate the online sensor weekly and calibrate the machine vision system monthly to ensure the accuracy and reliability of the monitoring data and provide reliable data support for the dipping glue quality control.

[0032] Based on the above embodiments of the present invention, relevant personnel can make various changes and modifications without departing from the technical concept of the present invention through the above description. The technical scope of the present invention is not limited to the content in the specification, and its technical scope must be determined according to the scope of the claims.

Claims

1. A method for controlling the amount of glue impregnation of an alkali-resistant glass fiber mesh cloth, characterized in that: The method comprises: The surface morphology structural parameters and chemical property parameters of the glass fiber mesh cloth are obtained, and the wettability and affinity of the fiber surface are judged according to the preset threshold value of the dipping process parameters, and the initial dipping amount range is determined; plasma treatment is performed on the surface of the glass fiber mesh cloth to introduce hydrophilic groups, increase the fiber surface energy, enhance the affinity with the glue solution, and achieve high dipping sufficiency under low dipping amount; according to the structural parameters of the glass fiber mesh cloth such as fiber diameter and fiber spacing, the finite element analysis method is used to simulate the spreading state of the glue solution on the fiber surface and the penetration depth inside the fiber bundle under different dipping amounts, and the quantitative relationship between the dipping amount and the dipping sufficiency is obtained; based on the above quantitative relationship between the dipping amount and the dipping sufficiency, by changing the type and amount of additives in the glue solution formula, the viscosity and surface tension of the glue solution are adjusted, and the change law of the uniformity of the glue solution distribution under different dipping amounts is obtained, and the optimal glue solution formula is determined in combination with the requirements of dipping sufficiency, and the uniformity of the glue solution distribution is quantitatively measured by optical microscope observation and image analysis software. The coverage rate and thickness distribution of the adhesive on the fiber surface are used to describe it; for the determined optimal adhesive formula and different dipping amounts, the orthogonal experimental design method is used to optimize the process parameters such as dipping time and number of dipping times, and the best matching scheme between the dipping amount and the process parameters is screened out through tensile strength, alkali resistance, bending strength, and shear strength performance tests; a nonlinear prediction model between the dipping amount, adhesive performance, fiber surface properties, dipping process parameters and dipping effect is established by combining numerical simulation with machine learning. The optimal combination of various parameters is found through the optimization algorithm to guide the precise control of the dipping amount. The dipping effect includes the adequacy of dipping, uniformity of adhesive distribution, internal penetration depth of the fiber bundle and the mechanical properties of the final composite material; in actual production, the dipping amount and process parameters are set according to the prediction model, and the real-time feedback control of the dipping amount is achieved through online monitoring of parameters such as adhesive viscosity and surface tension, combined with the detection of the surface morphology of the mesh cloth after dipping by machine vision, to ensure the stability of the dipping quality.

2. The method for controlling the amount of alkali-resistant glass fiber mesh cloth impregnation according to claim 1, characterized in that: The method of obtaining the surface morphology structure parameters and chemical property parameters of the glass fiber mesh cloth, judging the wettability and affinity of the fiber surface according to the preset dipping process parameter threshold, and determining the initial dipping amount range includes: The surface morphology image of the glass fiber mesh cloth was obtained by scanning electron microscopy, and the structural parameters of the fiber surface, including fiber diameter and porosity, were extracted using image processing algorithms. X-ray photoelectron spectrometer is used to test the chemical properties of glass fiber surface, including surface element composition and functional group types; According to the preset dip process parameter threshold, it is judged whether the fiber surface structure parameters and chemical property parameters meet the requirements. If they do, it is determined that the fiber surface has good wettability and resin affinity; According to the wettability and affinity of the fiber surface and the dipping process parameters, a dipping amount prediction model was established through the support vector machine regression algorithm to obtain the reasonable range of the initial dipping amount. If the actual dipping amount is lower than the lower limit of the predicted range, the dipping amount is increased by increasing the squeezing force of the dipping roller; If the actual dipping amount is higher than the upper limit of the predicted range, the dipping amount is reduced by reducing the squeezing force of the dipping roller; During the dipping process, the dipping amount is monitored in real time through machine vision, and the monitoring data is input into the dipping amount prediction model for comparison, and the dipping process parameters are dynamically adjusted to ensure that the dipping amount is always within a reasonable range; Genetic algorithm is used to optimize the dipping process parameters, with the stability of dipping amount and glass fiber performance as the optimization goals, to obtain the best combination of dipping process parameters to guide production practice.

3. The method for controlling the amount of alkali-resistant glass fiber mesh cloth impregnation according to claim 1, characterized in that: The plasma treatment is performed on the surface of the glass fiber mesh cloth to introduce hydrophilic groups, increase the fiber surface energy, enhance the affinity with the glue solution, and achieve high glue sufficiency at a low glue dipping amount, including: Obtain a glass fiber mesh cloth, place it in a plasma treatment device for surface treatment, introduce hydrophilic groups to the fiber surface through plasma, and increase the fiber surface energy; According to the material and structural characteristics of the glass fiber mesh cloth, determine the appropriate plasma treatment process parameters, including power, pressure and gas composition, so as to introduce sufficient hydrophilic groups on the fiber surface while avoiding damage to the fiber; Performing a surface energy test on the glass fiber mesh cloth after plasma treatment to determine whether it meets the preset hydrophilicity requirements. If not, returning to step 2 to adjust the process parameters and re-processing; The hydrophilized glass fiber mesh is immersed in the adhesive solution, and the adhesive solution is fully infiltrated and adhered to the fiber surface under the condition of low dipping amount through the affinity between the fiber surface and the adhesive solution; Use machine vision technology to inspect the glass fiber mesh after dipping, obtain the distribution of the glue on the fiber surface, and determine whether the uniformity and sufficiency of the dipping meet the requirements; If the dipping quality does not meet the standard, the affinity between the glue and the hydrophilized fiber surface can be improved by optimizing the glue formula and adjusting the dipping process parameters, such as dipping time and glue viscosity, to improve the dipping effect. The glass fiber mesh cloth that has passed the impregnation test is sent to the drying equipment and dried under controlled temperature and time conditions to solidify the glue and firmly adhere to the fiber surface to obtain a reinforced glass fiber mesh cloth product with high impregnation sufficiency.

4. The method for controlling the amount of alkali-resistant glass fiber mesh cloth impregnation according to claim 1, characterized in that: According to the structural parameters of the glass fiber mesh cloth such as fiber diameter and fiber spacing, the finite element analysis method is used to simulate the spreading state of the glue on the fiber surface and the penetration depth inside the fiber bundle under different dipping amounts, and obtain the quantitative relationship between the dipping amount and the dipping adequacy, including: According to the structural parameters of the glass fiber mesh cloth, such as the fiber diameter and fiber spacing, a finite element analysis model is established; Set different dipping amounts to simulate the spreading state of the glue on the fiber surface; If the glue does not completely cover the fiber surface, increase the amount of glue and re-simulate the glue spreading state until the glue completely covers the fiber surface; According to the simulation results, the spreading area of ​​the adhesive on the fiber surface under different amounts of adhesive is obtained; Simulate the penetration depth of the adhesive into the fiber bundle under different amounts of adhesive; If the penetration depth does not reach the preset threshold, the dipping amount is increased and the penetration depth is re-simulated until the preset threshold is reached; Through multiple simulations, the spreading area of ​​the adhesive on the fiber surface and the penetration depth inside the fiber bundle under different impregnation amounts were obtained, and a quantitative relationship model between the impregnation amount and the sufficiency of impregnation was established.

5. The method for controlling the amount of alkali-resistant glass fiber mesh cloth impregnation according to claim 1, characterized in that: Based on the quantitative relationship between the amount of dipping and the adequacy of dipping, the viscosity and surface tension of the glue solution are adjusted by changing the type and amount of the auxiliary agent in the glue solution formula, and the change law of the uniformity of the glue solution distribution under different dipping amounts is obtained. Combined with the requirements of dipping adequacy, the optimal glue solution formula is determined; the uniformity of the glue solution distribution is described by optical microscope observation and image analysis software to quantitatively calculate the coverage and thickness distribution of the glue solution on the fiber surface, including: According to the preset glue formula, multiple groups of glue samples with different types and dosages of additives are obtained, the viscosity and surface tension parameters of each sample are measured, and the correlation between the additive parameters and the physical property parameters of the glue is obtained; For different amounts of adhesive, the distribution of multiple adhesive samples on the fiber surface was observed using an optical microscope, and the adhesive coverage and thickness distribution parameters were calculated using image analysis software to establish a quantitative relationship model between the amount of adhesive and the uniformity of adhesive distribution. According to the target requirements of glue adequacy, the qualified range of glue coverage and thickness distribution is determined, which is used as a constraint condition. The optimal glue amount that meets the constraints is solved through an optimization algorithm by combining the glue amount and distribution uniformity model. According to the optimal dipping amount, combined with the correlation between the auxiliary parameters and the physical properties of the glue solution, a glue solution formula optimization model is established using machine learning algorithms, such as support vector machines or neural networks, to predict and select the optimal glue solution formula; For the preferred glue formula, a verification experiment is carried out, and the glue coverage and thickness distribution are calculated using optical microscope observation and image analysis software to determine whether it meets the requirements for glue adequacy; If the requirements are met, the formula is determined to be the best glue formula; If the requirements are not met, return to step 4, adjust the optimization model or constraint conditions, and re-optimize the formula until the requirements are met; According to the optimal glue formula, an appropriate amount of glue is prepared for the subsequent fiber dipping process to ensure the dipping quality and efficiency and improve the performance stability of fiber-reinforced composite materials.

6. The method for controlling the amount of glue impregnation of alkali-resistant glass fiber mesh cloth according to claim 1, characterized in that: The orthogonal test design method is used to optimize the process parameters such as the dipping time and the number of dipping times for the determined optimal glue solution formula and different dipping amounts, and the best matching scheme between the dipping amount and the process parameters is screened out through tensile strength, alkali resistance, bending strength, and shear strength performance tests, including: According to the determined optimal glue solution formula, different levels of dipping glue are set, and the dipping time and dipping times under each level are determined by using the orthogonal experimental design method; For each orthogonal test scheme, the corresponding dipped specimens were prepared and the tensile strength, alkali resistance, flexural strength and shear strength performance tests were carried out; According to the performance test results, the variance analysis method is used to determine the significance of the impact of each process parameter on the performance index, and to determine the optimal level combination of each parameter; Through multi-objective optimization algorithm, various performance indicators are comprehensively considered to obtain the best matching solution for dipping time and dipping times under different dipping amounts; Conduct verification tests on the best matching solution to test whether its performance indicators meet the requirements. If they do, it will be determined as the final process parameter combination; According to the final determined dipping amount and process parameters, establish dipping process specifications, specify the key points and precautions for each process, so as to ensure the stability of the production process and the consistency of product quality; Online monitoring technology is used to collect key parameters such as temperature, time and dipping amount during the dipping process in real time. Statistical process control methods are used to promptly detect and correct process deviations and continuously optimize the dipping process.

7. The method for controlling the amount of glue impregnation of alkali-resistant glass fiber mesh cloth according to claim 1, characterized in that: The method of combining numerical simulation with machine learning is used to establish a nonlinear prediction model between the amount of dipping, the performance of the glue solution, the surface properties of the fiber, the dipping process parameters and the dipping effect, and the optimal combination of the parameters is found through the optimization algorithm to guide the precise control of the dipping amount; the dipping effect includes the adequacy of dipping, the uniformity of the glue distribution, the penetration depth inside the fiber bundle and the mechanical properties of the final composite material, including: Obtain the dipping process parameters, adhesive performance parameters and fiber surface property parameters, and establish the corresponding mathematical model; The numerical simulation method is used to solve the established mathematical model and obtain the physical field distribution of the dipping process; According to the numerical simulation results, the characteristic parameters related to the dipping effect are extracted, such as dipping sufficiency, glue distribution uniformity and penetration depth inside the fiber bundle. Obtain a certain number of dipping process parameter combinations and corresponding dipping effect test data as training data sets for machine learning models; Use machine learning algorithms such as support vector machines or neural networks to establish a nonlinear prediction model between dipping process parameters and dipping effects; Using optimization algorithms such as genetic algorithms, taking the dipping effect as the optimization goal, we can find the optimal combination of dipping process parameters and achieve precise control of the dipping amount. According to the optimized process parameter combination, the dipping process was optimized, composite material samples were prepared and their mechanical properties were tested to verify the effectiveness of dipping amount control.

8. The method for controlling the amount of glue impregnation of alkali-resistant glass fiber mesh cloth according to claim 1, characterized in that: In actual production, the dipping amount and process parameters are set according to the prediction model, and the viscosity of the glue solution, surface tension and other parameters are monitored online, and the surface morphology of the mesh cloth after dipping is detected by machine vision to achieve real-time feedback control of the dipping amount and ensure the stability of the dipping quality, including: According to the prediction model, the initial dipping process parameters, including dipping time, dipping speed and glue solution temperature, are set as the benchmark values ​​for dipping control; Use online sensors to monitor glue performance indicators in real time, including glue viscosity and surface tension, and obtain real-time data on glue performance; The machine vision system is used to collect and analyze the surface morphology of the mesh cloth after dipping, extract the surface morphology features, and determine whether the dipping amount is uniform; Input the real-time monitoring glue performance data and the surface morphology characteristics analyzed by machine vision into the prediction model, dynamically adjust the dipping process parameters, and realize online feedback control of the dipping amount; If the glue performance monitored in real time exceeds the preset threshold range, or the machine vision detects uneven surface morphology, the early warning mechanism is triggered to automatically adjust the dipping process parameters to ensure that the dipping quality meets the requirements; Continuously record and analyze production process data, including performance trend of adhesive solution and distribution of surface morphology, optimize prediction models and process parameters, and improve the accuracy and stability of adhesive dipping control; Regularly calibrate and maintain online sensors and machine vision systems to ensure the accuracy and reliability of monitoring data and provide reliable data support for dipping quality control.

Images