Glass fiber production process processing method and system based on machine learning

Abstract

The invention discloses a glass fiber production process processing method and system based on machine learning, and belongs to the technical field of electrical digital data process.The method comprises the following steps that glass fiber drawing effect parameters are collected and analyzed to obtain glass fiber drawing effect indexes; analyzing based on the glass fiber wire drawing effect index, judging whether wire drawing cooling treatment is executed or not, if the wire drawing cooling treatment is executed, sending a cooling signal to cooling equipment, and if the wire drawing cooling treatment is not executed, adjusting the wire drawing equipment; cooling equipment aging indexes are obtained through analysis, aging correction factors are obtained through matching, cooling space environment influence parameters are synchronously obtained, cooling space environment influence indexes are obtained through analysis, and temperature correction factors are obtained through matching; according to the method, the cooling parameters of the cooling equipment are preliminarily corrected, the corrected adjusting parameters of the cooling equipment are obtained, glass fibers are cooled, and the problem that in the prior art, the quality of the glass fibers is not stable is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic digital data processing, and in particular to a glass fiber production process processing method and system based on machine learning. Background Art

[0002] With the development of industry, glass fiber has received great attention and has been applied in many fields such as aerospace, construction and building materials. The existing glass fiber system is achieved by obtaining the glass fiber cloth sintering control parameters to predict temperature information or perform structural mechanics simulation.

[0003] For example, the invention patent with publication number CN117540177B discloses a method and system for detecting and optimizing the temperature of glass fiber cloth out of the furnace, which includes: collecting multiple temperature detection information; obtaining glass fiber cloth simmering control parameters and generating multiple temperature prediction information; generating deviated particle coarseness and discrete particle coarseness; generating a first judgment result and a second judgment result; when the first judgment result is infeasible, or / and the second judgment result is infeasible, performing error compensation control on the glass fiber cloth simmering control parameters, and at the same time, performing continuous temperature detection and optimization.

[0004] However, in the process of implementing the technical solutions of the invention in the embodiments of the present application, the present application found that the above technology has at least the following technical problems:

[0005] In the existing technology, temperature information is mainly collected and analyzed in combination with the smoldering control parameters of the glass fiber cloth. However, these technical methods mainly rely on model prediction and real-time temperature control system adjustment. The existing technology lacks refined monitoring of the glass fiber drawing process during the production process. In the actual glass fiber manufacturing process, the impact of problems such as unstable drawing speed or insensitive drawing device on the glass fiber drawing quality is ignored, resulting in the problem of unstable glass fiber quality. Summary of the Invention

[0006] The embodiments of the present application solve the problem of unstable glass fiber quality in the prior art by providing a glass fiber production process processing method and system based on machine learning, and achieve precise control of glass fiber quality.

[0007] An embodiment of the present application provides a glass fiber production process processing method based on machine learning, comprising the following steps: when a drawing device receives a drawing signal, it starts to execute glass fiber drawing production processing, synchronously collects glass fiber drawing effect parameters, and analyzes to obtain a glass fiber drawing effect index; based on the glass fiber drawing effect index, it is analyzed to determine whether to execute drawing cooling processing, and if the drawing cooling processing is executed, a cooling signal is sent to the cooling device; if the drawing cooling processing is not executed, the drawing device is adjusted; when the cooling device receives a cooling signal, it obtains the cooling execution device aging parameter, and analyzes to obtain the cooling device aging index, thereby matching to obtain an aging correction coefficient, synchronously obtains the cooling space environment impact parameter, and analyzes to obtain the cooling space environment impact index, thereby matching to obtain a temperature correction factor; based on the aging correction coefficient and the temperature correction factor, the cooling parameter of the cooling device is preliminarily corrected to obtain the corrected cooling device adjustment parameter, and the glass fiber cooling processing is performed based on the corrected cooling device adjustment parameter.

[0008] Furthermore, the collecting of glass fiber drawing effect parameters and the analysis to obtain glass fiber drawing effect indicators specifically include: randomly arranging sampling points for each glass fiber, using a camera to monitor the glass fiber drawing process in real time, and collecting glass fiber drawing effect parameters, the collected glass fiber drawing effect parameters include the number of cracks of each glass fiber and the color, refractive index and temperature of each glass fiber sampling point; obtaining a preset drawing effect standard set in a database, and analyzing it with the glass fiber drawing effect parameters to obtain a glass fiber drawing effect indicator; the drawing effect standard set includes standard values for the number of cracks, standard values for color, standard values for refractive index and standard values for temperature; the glass fiber drawing effect indicator is used to characterize the production effect of glass fiber drawing.

[0009] Furthermore, the analysis is performed based on the glass fiber drawing effect index to determine whether to perform the drawing cooling treatment. The specific steps include: obtaining the glass fiber drawing effect threshold preset in the database, and comparing it with the glass fiber drawing effect index. If the glass fiber drawing effect index is above the glass fiber drawing effect threshold, the drawing cooling treatment is performed; if the glass fiber drawing effect index is less than the glass fiber drawing effect threshold, the drawing cooling treatment is not performed.

[0010] Furthermore, when the cooling device receives the cooling signal, it obtains the cooling execution device aging parameters, and analyzes them to obtain the cooling device aging index. The specific steps include: when the cooling device receives the cooling signal, it obtains the cooling execution device aging parameters, and the cooling execution device aging parameters include: the maximum coolant flow rate, the maximum coolant flow rate, the maximum vibration amplitude of the cooling device and the maximum coolant temperature deviation within the preset inspection period; obtains the cooling execution device reference set preset in the database, and analyzes it with the cooling execution device aging parameters to obtain the cooling device aging index; the cooling execution device reference set includes the historical maximum coolant flow rate, the historical maximum coolant flow rate, the cooling device vibration amplitude reference value and the coolant temperature deviation allowable value; the cooling device aging index is used to characterize the aging degree of the cooling device.

[0011] Furthermore, the aging correction coefficient is obtained by matching, and the specific steps include: obtaining the aging index intervals of each cooling device preset in the database and the aging reference correction coefficient corresponding to each cooling device aging index interval, comparing the cooling device aging index with each cooling device aging index interval, and if the cooling device aging index is within a certain cooling device aging index interval, obtaining the aging reference correction coefficient corresponding to the cooling device aging index interval as the aging correction coefficient.

[0012] Furthermore, the cooling space environmental impact index is obtained and the temperature correction factor is obtained by matching. The specific steps include: obtaining the volume of the cooling space, matching the volume of the cooling space with each cooling volume interval in the database, obtaining the number of cooling point layouts in the cooling space, and arranging the cooling points to obtain the air flow rate, ambient temperature and ambient humidity of each cooling layout point; obtaining the cooling space reference impact set preset in the database, and analyzing it with the cooling space environmental impact parameters to obtain the cooling space environmental impact index; the cooling space reference impact set includes an air flow rate reference value, an ambient temperature reference value and an ambient humidity reference value; the cooling space environmental impact index is used to characterize the degree of influence of the environmental factors of the cooling space on the cooling effect of the cooling equipment; obtaining each cooling space environmental impact index interval preset in the database and the temperature reference correction factor corresponding to each cooling space environmental impact index interval, and comparing them with the cooling space environmental impact index. If the cooling space environmental impact index is within a certain cooling space environmental impact index interval, then obtaining the temperature reference correction factor corresponding to the cooling space environmental impact index interval as the temperature correction factor.

[0013] Furthermore, the cooling space environmental impact index is obtained by:

[0014]

[0015] Where, LK represents the cooling space environmental impact index, LSk Indicates the air flow rate of the kth cooling layout point, k represents the number of the cooling layout point, k = 1, 2, 3, ..., k max , k max Indicates the total number of cooling arrangement points, ΔLS indicates the reference value of air flow rate in the cooling area, LH k represents the ambient temperature of the kth cooling layout point, ΔLH represents the ambient temperature reference value, and LD k represents the ambient humidity of the kth cooling layout point, and ΔLD represents the ambient humidity reference value.

[0016] Furthermore, the cooling parameters of the cooling equipment are preliminarily corrected based on the aging correction coefficient and the temperature correction factor to obtain corrected cooling equipment adjustment parameters. The specific steps include: obtaining the cooling parameters of the cooling equipment, the cooling parameters of the cooling equipment include cooling water flow rate, cooling water temperature and cooling water pressure; preliminarily correcting the cooling water flow rate, cooling water pressure and cooling air flow rate of the cooling equipment based on the aging correction coefficient, and preliminarily correcting the cooling water temperature and cooling air temperature of the cooling equipment based on the temperature correction factor to obtain corrected cooling equipment adjustment parameters; the corrected cooling equipment adjustment parameters include the preliminarily corrected cooling water flow rate, cooling water temperature and cooling water pressure.

[0017] Furthermore, it also includes obtaining the actual glass fiber quality indicators after the glass fiber cooling treatment, and analyzing them to obtain the glass fiber quality judgment results, and judging whether to input them as verification set data into the machine learning quality prediction model based on the glass fiber quality judgment results, thereby updating the machine learning quality prediction model.

[0018] The embodiment of the present application provides a glass fiber production process processing system based on machine learning, including: a wire drawing execution analysis module, a wire drawing cooling judgment module, a correction processing module and a preliminary correction processing module; wherein, the wire drawing execution analysis module is used to start the glass fiber drawing production processing after the wire drawing equipment receives the wire drawing signal, synchronously collect the glass fiber drawing effect parameters, and analyze to obtain the glass fiber drawing effect index; the wire drawing cooling judgment module is used to analyze based on the glass fiber drawing effect index to determine whether to execute the wire drawing cooling process, and if the wire drawing cooling process is executed, a cooling signal is sent to the cooling device, and if not, When performing wire drawing cooling treatment, the wire drawing equipment is adjusted; the correction processing module is used to obtain the aging parameters of the cooling execution equipment after the cooling equipment receives the cooling signal, and analyze the aging index of the cooling equipment, thereby matching the aging correction coefficient, and simultaneously obtain the cooling space environment impact parameter, and analyze the cooling space environment impact index, thereby matching the temperature correction factor; the preliminary correction processing module is used to perform preliminary correction on the cooling parameters of the cooling equipment based on the aging correction coefficient and the temperature correction factor, obtain the corrected cooling equipment adjustment parameters, and perform glass fiber cooling treatment based on the corrected cooling equipment adjustment parameters.

[0019] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0020] 1. The glass fiber production process processing method based on machine learning provided by the present invention obtains the glass fiber drawing effect parameter analysis to obtain the glass fiber drawing effect index, and compares it with the glass fiber drawing effect threshold, so as to dynamically adjust the drawing equipment operating parameters, and then perform the drawing cooling treatment, and correct the cooling equipment adjustment parameters, which effectively solves the problem of unstable glass fiber quality in the prior art.

[0021] 2. The present invention obtains the aging index of the cooling equipment and matches the aging correction coefficient. At the same time, it combines the cooling space environmental impact parameters to analyze the cooling space environmental impact index and matches the temperature correction factor. The cooling parameters of the cooling equipment are corrected based on the aging correction coefficient and the temperature correction factor, thereby optimizing the operating performance of the cooling equipment and improving the quality of the glass fiber.

[0022] 3. By inputting the glass fiber drawing effect parameters and the corrected cooling equipment adjustment parameters into the machine learning quality prediction model and comparing them with the actual glass fiber quality indicators, the model is dynamically updated to achieve accurate prediction of glass fiber quality and optimization of process parameters, thereby achieving optimized processing of the glass fiber production process. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 A flow chart of a glass fiber production process based on machine learning provided in an embodiment of the present application;

[0024] Figure 2 A schematic diagram of the structure of a glass fiber production process processing system based on machine learning provided in an embodiment of the present application. DETAILED DESCRIPTION

[0025] The embodiments of the present application solve the problem of unstable glass fiber quality in the prior art by providing a glass fiber production process processing method and system based on machine learning, thereby achieving precise control of glass fiber quality.

[0026] In order to better understand the above technical solution, the above technical solution will be described in detail below with reference to the accompanying drawings and specific implementation methods.

[0027] like Figure 1 As shown, it is a flow chart of a glass fiber production process processing method based on machine learning provided in an embodiment of the present application. The method is applied to a glass fiber production process processing system based on machine learning, and the method includes the following steps: when the drawing device receives a drawing signal, it starts to execute glass fiber drawing production processing, synchronously collects glass fiber drawing effect parameters, and analyzes to obtain glass fiber drawing effect indicators; based on the glass fiber drawing effect indicators, it is analyzed to determine whether to execute drawing cooling processing. If the drawing cooling processing is executed, a cooling signal is sent to the cooling device; if the drawing cooling processing is not executed, the drawing device is adjusted; when the cooling device receives a cooling signal, it obtains the aging parameters of the cooling execution device, and analyzes to obtain the aging index of the cooling device, thereby matching to obtain the aging correction coefficient, synchronously obtains the cooling space environment impact parameters, and analyzes to obtain the cooling space environment impact index, thereby matching to obtain the temperature correction factor; based on the aging correction coefficient and the temperature correction factor, the cooling parameters of the cooling device are preliminarily corrected to obtain the corrected cooling device adjustment parameters, and the glass fiber cooling processing is performed based on the corrected cooling device adjustment parameters.

[0028] In this embodiment, when the drawing equipment receives the drawing signal, it starts to execute the glass fiber drawing production process, synchronously collects the glass fiber drawing effect parameters, and analyzes to obtain the glass fiber drawing effect index, thereby realizing real-time monitoring and control of the glass fiber drawing process; by judging whether to execute the drawing cooling process based on the glass fiber drawing effect index, and sending a cooling signal or adjusting the drawing equipment, the coordination of the drawing and cooling processes is optimized; by obtaining the aging parameters of the cooling execution equipment and the cooling space environmental influencing parameters, the aging correction coefficient and the temperature correction factor are matched, and the glass fiber is cooled based on the corrected cooling equipment adjustment parameters, thereby realizing intelligent control and precise adjustment of the glass fiber drawing and cooling processes, effectively solving the problem of unstable glass fiber quality in the prior art.

[0029] Furthermore, glass fiber drawing effect parameters are collected and analyzed to obtain glass fiber drawing effect indicators, specifically including: randomly arranging sampling points for each glass fiber, using a camera to monitor the glass fiber drawing process in real time, and collecting glass fiber drawing effect parameters, including the number of cracks in each glass fiber and the color, refractive index and temperature of each glass fiber sampling point; obtaining a preset drawing effect standard set in a database, and analyzing it with the glass fiber drawing effect parameters to obtain a glass fiber drawing effect indicator; the drawing effect standard set includes standard values for the number of cracks, standard values for color, standard values for refractive index and standard values for temperature; the glass fiber drawing effect indicator is used to characterize the production effect of glass fiber drawing.

[0030] In this embodiment, the color of the glass fiber sampling point refers to the color quantified using the RGB color model, specifically the R value. It should be noted that the glass fiber drawing effect parameters are obtained by taking pictures or videos with a camera and then identifying the taken pictures or videos. For example, the number of cracks can be obtained by taking pictures of the glass fiber with a camera and statistically analyzing the pictures using image processing software (such as Photoshop). The refractive index can be obtained by measuring with a refractometer, and the temperature can be obtained by using thermal imaging technology, where the temperature refers to the surface temperature of the glass fiber during the drawing process, and the temperature can also be obtained by a temperature sensor.

[0031] The fiberglass drawing effect index is derived by analyzing the fiberglass drawing effect parameters, taking into account the interplay between these parameters. For example, excessively high temperatures can cause structural instability in the fiberglass, leading to cracks. The appearance of cracks, in turn, affects the fiberglass's refractive index. Color changes are caused by temperature fluctuations and unstable drawing speeds. Excessively high temperatures can cause the fiberglass's color to deviate from the standard value, resulting in uneven color differences. High temperatures can also cause changes in the fiberglass's refractive index.

[0032] The fewer cracks, the more complete the glass fiber structure and the better the drawing effect. The closer the refractive index is to the standard refractive index value, the more stable the glass fiber, the better the glass fiber quality and the greater the drawing effect index. The closer the temperature at the glass fiber sampling point is to the standard temperature value, the more stable the temperature is and the more stable the quality of the resulting glass fiber.

[0033] The specific method to obtain the glass fiber drawing effect index is as follows:

[0034]

[0035] It should be noted that XG represents the glass fiber drawing effect index, e represents the natural constant, and XZ i Indicates the number of cracks in the i-th glass fiber, i represents the number of the glass fiber, i=1,2,...,i max ,i max Indicates the total number of glass fibers, ΔXZ indicates the standard value of crack number, XD j Indicates the color of the jth glass fiber sampling point, j represents the number of the sampling point, j=1,2,...,j max ,j max Indicates the total number of sampling points, ΔXD indicates the color standard value, XQ j represents the refractive index of the jth glass fiber sampling point, ΔXQ represents the standard value of the refractive index, and XC j represents the temperature of the j-th glass fiber sampling point, and ΔXC represents the temperature standard value.

[0036] By acquiring the operating parameters of the drawing equipment and simultaneously collecting the glass fiber drawing effect parameters, it is possible to monitor the changes in various key indicators during the drawing process in real time, thereby achieving dynamic tracking and precise analysis of the drawing process, thereby improving the drawing quality and production efficiency of the glass fiber. By evaluating the relationship between the glass fiber drawing effect parameters and the standard values, it is possible to promptly detect deviations between the drawing parameters and the standard values, and through analysis, obtain the drawing effect indicators, thereby optimizing the operating parameters of the drawing equipment and making the drawing process more in line with process requirements. By synchronizing the operating parameters of the drawing equipment and the glass fiber drawing effect parameters, real-time monitoring and precise analysis of the drawing process are achieved; thereby optimizing the matching of process parameters with standard values, improving the process accuracy and product quality consistency of the glass fiber drawing process, and achieving the goal of enhancing process stability and improving production efficiency.

[0037] Furthermore, based on the glass fiber drawing effect index, analysis is performed to determine whether to perform drawing cooling treatment. The specific steps include: obtaining the glass fiber drawing effect threshold preset in the database, and comparing it with the glass fiber drawing effect index. If the glass fiber drawing effect index is above the glass fiber drawing effect threshold, the drawing cooling treatment is performed. If the glass fiber drawing effect index is less than the glass fiber drawing effect threshold, the drawing cooling treatment is not performed.

[0038] In this embodiment, by determining whether the glass fiber drawing effect index meets the threshold requirements, the drawing cooling process is performed only when necessary, avoiding the need for cooling treatment even though the drawing process is unqualified, avoiding excessive use of cooling equipment, thereby reducing energy consumption and improving the efficiency of cooling resource utilization. By dynamically adjusting the cooling process based on the glass fiber drawing effect index, the processing steps can be optimized according to the actual situation during the drawing process, ensuring the necessity and appropriateness of the cooling process, thereby effectively improving the quality stability and consistency of the glass fiber products. By reducing unnecessary cooling processes, the operating frequency and workload of the cooling equipment are reduced, thereby effectively extending the service life of the equipment.

[0039] Furthermore, when the cooling device receives the cooling signal, the cooling execution device aging parameters are obtained, and the cooling device aging index is obtained by analysis. The specific steps include: when the cooling device receives the cooling signal, the cooling execution device aging parameters are obtained, and the cooling execution device aging parameters include: the maximum coolant flow rate, the maximum coolant flow rate, the maximum vibration amplitude of the cooling device and the maximum coolant temperature deviation within the preset inspection period; the cooling execution device reference set preset in the database is obtained, and the cooling execution device aging parameters are analyzed to obtain the cooling device aging index; the cooling execution device reference set includes the historical maximum coolant flow rate, the historical maximum coolant flow rate, the cooling device vibration amplitude reference value and the coolant temperature deviation allowable value; the cooling device aging index is used to characterize the aging degree of the cooling device.

[0040] In this embodiment, the maximum coolant flow rate can be measured using an ultrasonic flowmeter or laser Doppler velocimeter installed in the cooling pipeline, or can be measured using a tachometer. The maximum coolant flow rate can be measured using a volumetric flow meter. The maximum vibration amplitude of the cooling device can be obtained by using a vibration sensor to record the maximum vibration amplitude of the cooling device in real time during operation. The coolant temperature deviation can be measured and statistically obtained using a temperature sensor.

[0041] The cooling equipment aging index is derived by analyzing cooling equipment aging parameters (maximum coolant flow rate, maximum coolant flow rate, maximum cooling equipment vibration amplitude, and coolant temperature deviation). This takes into account the interplay between these parameters. For example, a slower maximum coolant flow rate results in slower cooling, which reduces the overall cooling efficiency of the cooling equipment. Slower coolant flow rate and coolant flow rate indicate more severe cooling equipment aging, resulting in greater cooling equipment instability and, consequently, increased cooling equipment vibration amplitude. Greater coolant temperature deviation indicates more unstable cooling equipment operation, indicating more severe cooling equipment aging.

[0042] The aging index of the cooling equipment is obtained by:

[0043]

[0044] Where LH represents the cooling equipment aging index, FG represents the maximum coolant flow rate, ΔFG represents the historical maximum coolant flow rate, FD represents the maximum coolant flow rate, ΔFD represents the historical maximum coolant flow rate, Fσ represents the maximum vibration amplitude of the cooling equipment, ΔFσ represents the reference value of the cooling equipment vibration amplitude, FL represents the maximum coolant temperature deviation, ΔFL represents the allowable coolant temperature deviation, and e represents a natural constant.

[0045] After the cooling device receives the cooling signal, the aging parameters of the cooling execution device are obtained and the cooling device aging index is analyzed to evaluate the operating status and aging degree of the cooling device, thereby providing an accurate correction basis for subsequent cooling operations. At the same time, by introducing the analysis of multiple parameters such as the cooling pipe pollutant coverage rate, the cooling equipment refrigeration efficiency, and the maximum flow rate of the coolant, the performance status of the equipment can be reflected in multiple dimensions, effectively improving the accuracy of the cooling equipment aging index. By comparing the aging parameters of the cooling execution device with the factory parameters, the corresponding aging range and correction coefficient can be quickly matched, thereby achieving precise adjustment of the cooling equipment operating parameters, significantly improving the operating efficiency and service life of the cooling equipment, and providing reliable protection for the glass fiber production process.

[0046] Furthermore, an aging correction coefficient is obtained by this matching, and the specific steps include: obtaining the aging index intervals of each cooling device preset in the database and the aging reference correction coefficient corresponding to each cooling device aging index interval, comparing the cooling device aging index with each cooling device aging index interval, and if the cooling device aging index is within a certain cooling device aging index interval, obtaining the aging reference correction coefficient corresponding to the cooling device aging index interval as the aging correction coefficient.

[0047] In this embodiment, by comparing the cooling equipment aging index with the cooling equipment aging index intervals preset in the database, the corresponding aging reference correction coefficient is accurately matched, thereby providing customized correction parameters for different degrees of equipment aging. This improves the pertinence and accuracy of cooling parameter adjustments, avoiding the degradation of cooling performance caused by manual judgment errors or inaccurate corrections. In addition, by introducing a matching mechanism between aging intervals and corresponding correction coefficients, the complex parameter calculation process is simplified, the response efficiency is improved, and the service life of the cooling equipment is extended.

[0048] Furthermore, the cooling space environmental impact index is obtained, and the temperature correction factor is obtained by matching it. The specific steps include: obtaining the volume of the cooling space, matching the volume of the cooling space with each cooling volume interval in the database, obtaining the number of cooling point layouts in the cooling space, and arranging the cooling points to obtain the air flow rate, ambient temperature and ambient humidity of each cooling layout point; obtaining the cooling space reference impact set preset in the database, and analyzing it with the cooling space environmental impact parameters to obtain the cooling space environmental impact index; the cooling space reference impact set includes an air flow rate reference value, an ambient temperature reference value and an ambient humidity reference value; the cooling space environmental impact index is used to characterize the degree of influence of the environmental factors of the cooling space on the cooling effect of the cooling equipment; obtaining each cooling space environmental impact index interval preset in the database and the temperature reference correction factor corresponding to each cooling space environmental impact index interval, and comparing them with the cooling space environmental impact index. If the cooling space environmental impact index is within a certain cooling space environmental impact index interval, then obtaining the temperature reference correction factor corresponding to the cooling space environmental impact index interval as the temperature correction factor.

[0049] In this embodiment, the air velocity in the cooling area can be measured using a hot-wire anemometer, an ultrasonic anemometer, or an airflow meter. The ambient temperature in the cooling space can be measured using an infrared thermometer or a temperature and humidity sensor. The ambient humidity in the cooling space can be obtained by obtaining real-time humidity data from the temperature and humidity sensor.

[0050] Furthermore, the cooling space environmental impact index is obtained, and the specific method is as follows:

[0051]

[0052] Where, LK represents the cooling space environmental impact index, LS k Indicates the air flow rate of the kth cooling layout point, k represents the number of the cooling layout point, k = 1, 2, 3, ..., k max , k maxIndicates the total number of cooling arrangement points, ΔLS indicates the reference value of air flow rate in the cooling area, LH k represents the ambient temperature of the kth cooling layout point, ΔLH represents the ambient temperature reference value, and LD k represents the ambient humidity of the kth cooling layout point, and ΔLD represents the ambient humidity reference value.

[0053] In this embodiment, the cooling space environmental impact index is obtained by analyzing the cooling space environmental impact parameters (air flow rate, ambient temperature and ambient humidity of each cooling layout point), taking into account the mutual influence relationship between these parameters. For example, the air flow rate will affect the heat diffusion efficiency of each cooling layout point, and at the same time affect the temperature distribution of the cooling space. The ambient temperature of the cooling space directly determines the cooling efficiency, while the ambient humidity indirectly affects the cooling process by affecting the thermal conductivity and evaporation rate of the air. When the ambient temperature is high, the amount of heat that the cooling device needs to dissipate increases, resulting in increased cooling difficulty. The higher the ambient humidity, the greater the water vapor content in the air, which is less conducive to heat loss. Therefore, the higher the ambient temperature and the greater the ambient humidity, the slower the heat exchange in the environment, which affects the cooling efficiency.

[0054] By acquiring air velocity, ambient temperature, and humidity parameters at each cooling point within the cooling space and comparing them with reference values in a database, we can generate cooling space environmental impact indicators, thereby quantifying the degree to which environmental factors influence cooling equipment performance. This effectively reduces fluctuations in cooling equipment operation caused by environmental changes and improves cooling performance stability. By monitoring these parameters in real time and matching correction factors, we can achieve dynamic optimization and control of cooling equipment, improve cooling efficiency, simplify the cooling equipment parameter calibration process in complex environments, and reduce errors caused by manual intervention.

[0055] Furthermore, based on the aging correction coefficient and the temperature correction factor, the cooling parameters of the cooling equipment are preliminarily corrected to obtain the corrected cooling equipment adjustment parameters. The specific steps include: obtaining the cooling parameters of the cooling equipment, the cooling parameters of the cooling equipment include cooling water flow rate, cooling water temperature and cooling water pressure; based on the aging correction coefficient, the cooling water flow rate, cooling water pressure and cooling air flow rate of the cooling equipment are preliminarily corrected, and based on the temperature correction factor, the cooling water temperature and cooling air temperature of the cooling equipment are preliminarily corrected to obtain the corrected cooling equipment adjustment parameters; the corrected cooling equipment adjustment parameters include the preliminarily corrected cooling water flow rate, cooling water temperature and cooling water pressure.

[0056] In this embodiment, by making a preliminary correction to the cooling parameters of the cooling equipment based on the aging correction coefficient and the temperature correction factor, it is possible to effectively deal with the performance deviation of the cooling equipment caused by equipment aging and changes in the cooling environment during long-term operation. By making aging corrections to the cooling water flow rate, cooling water pressure and cooling air flow rate, it is possible to accurately compensate for changes in fluid dynamics parameters caused by equipment wear or equipment aging, thereby ensuring the stability of the cooling system. On the other hand, by correcting the cooling water temperature and cooling air temperature using the temperature correction factor, the cooling parameters can be dynamically adjusted to adapt to changes in ambient temperature, avoiding a reduction in cooling efficiency due to environmental factors. The corrected cooling equipment adjustment parameters lay the foundation for subsequent precise control, improve the operational adaptability of the cooling equipment under complex working conditions, thereby optimizing and improving the cooling performance and extending the service life of the equipment.

[0057] Furthermore, it also includes obtaining the actual glass fiber quality indicators after the glass fiber cooling treatment, and analyzing them to obtain the glass fiber quality judgment results, and judging whether to input them as verification set data into the machine learning quality prediction model based on the glass fiber quality judgment results, thereby updating the machine learning quality prediction model.

[0058] In this embodiment, it should be noted that the actual glass fiber quality index after the glass fiber is cooled is obtained and analyzed to obtain the glass fiber quality determination result. The specific steps include: obtaining the glass fiber quality determination threshold preset in the database, and comparing it with the actual glass fiber quality index after the glass fiber is cooled. If the actual glass fiber quality index after the glass fiber is cooled is above the glass fiber quality determination threshold, the glass fiber quality determination result is that the glass fiber quality is qualified; if the actual glass fiber quality index after the glass fiber is cooled is less than the glass fiber quality determination threshold, the glass fiber quality determination result is that the glass fiber quality is unqualified.

[0059] The glass fibers are sampled to obtain glass fiber samples, and the glass fiber samples are tested. The diameter range of the glass fiber refers to measuring the glass fiber sample diameter at both ends, and then calculating the diameter difference. All samples are analyzed, and the maximum diameter difference is marked as the diameter range of the glass fiber. The tensile strength, heat resistance temperature and number of surface cracks of the glass fiber refer to testing all glass fiber samples and then obtaining the average value.

[0060] The actual glass fiber quality index is obtained, and the steps include: obtaining quality parameters of the glass fiber, which include the diameter range, tensile strength, heat resistance temperature and the number of surface cracks of the glass fiber; obtaining a glass fiber benchmark set preset in a database, which includes a diameter range benchmark value, a tensile strength benchmark value, a heat resistance temperature benchmark value and a surface crack number benchmark value; and analyzing the quality parameters of the glass fiber with the glass fiber benchmark set to obtain the actual glass fiber quality index.

[0061] The diameter range of glass fibers can be measured using a laser diameter gauge. The tensile strength can be determined by randomly sampling the glass fibers, placing the sampled glass fiber samples in a tensile testing machine (such as an electronic universal testing machine) and stretching them, then recording the maximum tensile force at fiber breakage. The heat resistance temperature can be measured using a thermal analyzer, and the number of surface cracks can be determined by scanning and identifying them using a scanning electron microscope (SEM).

[0062] The actual glass fiber quality index is obtained by:

[0063]

[0064] Wherein, SZ represents the actual glass fiber quality index, XG represents the glass fiber drawing effect index, JC represents the diameter range of the glass fiber, ΔJC represents the diameter range reference value, SQ represents the tensile strength of the glass fiber, ΔSQ represents the tensile strength reference value, SW represents the heat resistance temperature of the glass fiber, ΔSW represents the heat resistance temperature reference value, SN represents the number of surface cracks of the glass fiber, ΔSN represents the surface crack number reference value, and e represents a natural constant.

[0065] The actual glass fiber quality index is obtained by analyzing the quality parameters of glass fiber (glass fiber diameter range, tensile strength, heat resistance temperature, maximum bending angle, and number of surface cracks). This takes into account the mutual influence between these parameters. For example, the smaller the diameter range, the better the geometric uniformity of the fiber, which can improve the tensile strength and heat resistance temperature, and reduce the formation of surface cracks. Fibers with higher tensile strength generally have better heat resistance. The higher the heat resistance temperature, the more conducive it is to improving the fiber's bending performance in high temperature environments, preventing the increase of cracks and the decrease of tensile properties.

[0066] The calculation formula of the machine learning quality prediction model is:

[0067]

[0068] Where, JQ represents the predicted quality value of glass fiber, XD j Indicates the color of the jth glass fiber sampling point, j represents the number of the sampling point, j=1,2,...,j max ,j max Indicates the total number of sampling points, ΔXD indicates the color standard value, XQ j represents the refractive index of the jth glass fiber sampling point, ΔXQ represents the standard value of the refractive index, and XC j represents the temperature of the j-th glass fiber sampling point, ΔXC represents the temperature standard value, FL represents the maximum deviation of the coolant temperature, ΔFL represents the allowable value of the coolant temperature deviation, and e represents a natural constant.

[0069] It should be noted that the quality prediction value of the glass fiber is obtained by obtaining the color, refractive index, and temperature of the glass fiber sampling point and performing a comprehensive analysis with the coolant temperature. This is to conduct timely monitoring and quality prediction during the glass fiber production process. If the predicted glass fiber quality is unqualified, it can be adjusted in time during the production process to prevent the phenomenon of unqualified product quality due to untimely adjustment, while avoiding waste of resources and improving production efficiency.

[0070] By analyzing the actual glass fiber quality indicators, the actual quality value of glass fiber production is obtained. According to the analysis of the actual glass fiber quality indicators, the machine learning quality prediction model is adjusted to obtain a more accurate glass fiber quality prediction value, thereby preventing the misjudgment of glass fiber quality prediction due to inaccurate glass fiber prediction values.

[0071] like Figure 2As shown, it is a structural diagram of the glass fiber production process processing system based on machine learning provided in an embodiment of the present application. The glass fiber production process processing system based on machine learning provided in an embodiment of the present application includes: a drawing execution analysis module, a drawing cooling judgment module, a correction processing module and a preliminary correction processing module; wherein the drawing execution analysis module is used to start the glass fiber drawing production processing after the drawing equipment receives the drawing signal, synchronously collect the glass fiber drawing effect parameters, and analyze them to obtain the glass fiber drawing effect index; the drawing cooling judgment module is used to analyze based on the glass fiber drawing effect index to determine whether to execute the drawing cooling treatment, and if the drawing cooling treatment is executed If the wire drawing cooling process is not performed, the wire drawing device is adjusted; the correction processing module is used to obtain the aging parameters of the cooling execution device after the cooling device receives the cooling signal, and analyze the aging index of the cooling device to obtain the aging correction coefficient, and simultaneously obtain the cooling space environment impact parameter, and analyze the cooling space environment impact index to obtain the temperature correction factor; the preliminary correction processing module is used to perform preliminary correction on the cooling parameters of the cooling device based on the aging correction coefficient and the temperature correction factor, obtain the corrected cooling device adjustment parameters, and perform glass fiber cooling process based on the corrected cooling device adjustment parameters.

[0072] To sum up, this embodiment obtains the glass fiber drawing effect index by obtaining the glass fiber drawing effect parameter analysis, and compares it with the glass fiber drawing effect threshold, so as to dynamically adjust the drawing equipment operating parameters, and then perform the drawing cooling process, and correct the cooling equipment adjustment parameters, which effectively solves the problem of unstable glass fiber quality in the existing technology.

[0073] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0074] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0075] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0076] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0077] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0078] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A glass fiber production process processing method based on machine learning, characterized in that: The following steps are involved: When the drawing equipment receives the drawing signal, it starts to execute the glass fiber drawing production process, synchronously collects the glass fiber drawing effect parameters, and analyzes them to obtain the glass fiber drawing effect index; Analyze the glass fiber drawing effect index to determine whether to perform the drawing cooling process. If the drawing cooling process is to be performed, send a cooling signal to the cooling device. If the drawing cooling process is not to be performed, adjust the drawing device. When the cooling device receives the cooling signal, it obtains the cooling execution device aging parameters, analyzes them to obtain the cooling device aging index, and matches them to obtain the aging correction coefficient. It also simultaneously obtains the cooling space environment impact parameters, analyzes them to obtain the cooling space environment impact index, and matches them to obtain the temperature correction factor. Based on the aging correction coefficient and the temperature correction factor, the cooling parameters of the cooling device are preliminarily corrected to obtain the corrected cooling device adjustment parameters, and the glass fiber cooling process is performed based on the corrected cooling device adjustment parameters.

2. The glass fiber production process method based on machine learning according to claim 1, characterized in that: The glass fiber drawing effect parameters are collected and analyzed to obtain glass fiber drawing effect indicators, specifically including: Sampling points are randomly arranged on each glass fiber, and the glass fiber drawing process is monitored in real time using a camera to collect glass fiber drawing effect parameters, including the number of cracks in each glass fiber and the color, refractive index and temperature of each glass fiber sampling point; Obtaining a preset drawing effect standard set in a database, and analyzing it with the glass fiber drawing effect parameters to obtain a glass fiber drawing effect index; The wire drawing effect standard set includes a standard value for the number of cracks, a standard value for color, a standard value for refractive index, and a standard value for temperature; The glass fiber drawing effect index is used to characterize the production effect of glass fiber drawing.

3. The glass fiber production process method based on machine learning according to claim 1, characterized in that: The analysis based on the glass fiber drawing effect index to determine whether to perform the drawing cooling process specifically includes the following steps: Obtain the glass fiber drawing effect threshold preset in the database and compare it with the glass fiber drawing effect index. If the glass fiber drawing effect index is above the glass fiber drawing effect threshold, the drawing cooling process is performed. If the glass fiber drawing effect index is less than the glass fiber drawing effect threshold, the drawing cooling process is not performed.

4. The glass fiber production process method based on machine learning according to claim 1, characterized in that: When the cooling device receives the cooling signal, it obtains the cooling execution device aging parameter and analyzes it to obtain the cooling device aging index. The specific steps include: When the cooling device receives the cooling signal, the cooling device aging parameters are obtained, wherein the cooling device aging parameters include: the maximum flow rate of the coolant, the maximum flow rate of the coolant, the maximum vibration amplitude of the cooling device, and the maximum deviation of the coolant temperature within a preset inspection period; Obtain a preset cooling execution equipment reference set in the database, and analyze it with the cooling execution equipment aging parameters to obtain the cooling equipment aging index; The cooling execution equipment reference set includes historical maximum coolant flow rate, historical maximum coolant flow rate, cooling equipment vibration amplitude reference value and coolant temperature deviation allowable value; The cooling equipment aging index is used to characterize the aging degree of the cooling equipment.

5. The glass fiber production process method based on machine learning according to claim 1, characterized in that: The aging correction coefficient is obtained by matching, and the specific steps include: Obtain the aging index intervals of each cooling device preset in the database and the aging reference correction coefficient corresponding to each cooling device aging index interval, compare the cooling device aging index with each cooling device aging index interval, and if the cooling device aging index is within a certain cooling device aging index interval, obtain the aging reference correction coefficient corresponding to the cooling device aging index interval as the aging correction coefficient.

6. The glass fiber production process method based on machine learning according to claim 1, characterized in that: The cooling space environment impact index is obtained, and the temperature correction factor is obtained by matching. Specifically, the steps include: Obtain the volume of the cooling space, match the volume of the cooling space with the cooling volume intervals in the database, obtain the number of cooling points in the cooling space, and then arrange the cooling points to obtain the air flow rate, ambient temperature, and ambient humidity of each cooling point; Obtain the cooling space reference impact set preset in the database, and analyze it with the cooling space environmental impact parameters to obtain the cooling space environmental impact index; The cooling space reference influence set includes an air flow rate reference value, an ambient temperature reference value, and an ambient humidity reference value; The cooling space environmental impact index is used to characterize the degree of influence of the environmental factors of the cooling space on the cooling effect of the cooling equipment; Obtain each cooling space environmental impact index interval preset in the database and the temperature reference correction factor corresponding to each cooling space environmental impact index interval, and compare them with the cooling space environmental impact index. If the cooling space environmental impact index is within a certain cooling space environmental impact index interval, obtain the temperature reference correction factor corresponding to the cooling space environmental impact index interval as the temperature correction factor.

7. The glass fiber production process method based on machine learning according to claim 6, characterized in that: The specific method for obtaining the cooling space environmental impact index is as follows: Where, LK represents the cooling space environmental impact index, LS k Indicates the air flow rate of the kth cooling layout point, k represents the number of the cooling layout point, k = 1, 2, 3, ..., k max , k max Indicates the total number of cooling arrangement points, ΔLS indicates the reference value of air flow rate in the cooling area, LH k represents the ambient temperature of the kth cooling layout point, ΔLH represents the ambient temperature reference value, and LD k represents the ambient humidity of the kth cooling layout point, and ΔLD represents the ambient humidity reference value.

8. The glass fiber production process method based on machine learning according to claim 1, characterized in that: The cooling parameters of the cooling device are preliminarily corrected based on the aging correction coefficient and the temperature correction factor to obtain the corrected cooling device adjustment parameters. The specific steps include: Acquiring cooling parameters of the cooling device, wherein the cooling parameters of the cooling device include cooling water flow rate, cooling water temperature and cooling water pressure; Perform preliminary correction processing on the cooling water flow rate, cooling water pressure and cooling air flow rate of the cooling equipment based on the aging correction coefficient, and perform preliminary correction processing on the cooling water temperature and cooling air temperature of the cooling equipment based on the temperature correction factor to obtain the corrected cooling equipment adjustment parameters; The corrected cooling equipment adjustment parameters include the initially corrected cooling water flow rate, cooling water temperature and cooling water pressure.

9. The glass fiber production process method based on machine learning according to claim 1, characterized in that: It also includes obtaining the actual glass fiber quality indicators after the glass fiber cooling treatment, and analyzing to obtain the glass fiber quality judgment results, and judging whether to input them as verification set data into the machine learning quality prediction model based on the glass fiber quality judgment results, thereby updating the machine learning quality prediction model.

10. A glass fiber production process processing system based on machine learning, characterized in that: include: Wire drawing execution analysis module, wire drawing cooling judgment module, correction processing module and preliminary correction processing module; The drawing execution analysis module is used to start the glass fiber drawing production process after the drawing equipment receives the drawing signal, synchronously collect the glass fiber drawing effect parameters, and analyze them to obtain the glass fiber drawing effect index; The drawing cooling judgment module is used to analyze the glass fiber drawing effect index and judge whether to perform the drawing cooling process. If the drawing cooling process is to be performed, a cooling signal is sent to the cooling device; if the drawing cooling process is not to be performed, the drawing device is adjusted; The correction processing module is used to obtain the cooling execution device aging parameters after the cooling device receives the cooling signal, and analyze the cooling device aging index to obtain the aging correction coefficient, and simultaneously obtain the cooling space environment impact parameters, and analyze the cooling space environment impact index to obtain the temperature correction factor; The preliminary correction processing module is used to perform preliminary correction on the cooling parameters of the cooling device based on the aging correction coefficient and the temperature correction factor to obtain the corrected cooling device adjustment parameters, and perform glass fiber cooling processing based on the corrected cooling device adjustment parameters.

Images