Intelligent monitoring method and system for high-performance fiber non-woven fabric production line

Abstract

The invention relates to the technical field of non-woven fabric production monitoring, and particularly discloses an intelligent monitoring method and system for a high-performance fiber non-woven fabric production line. The method mainly comprises the steps of obtaining a data set of a non-woven fabric production line; the fiber uniformity in the fiber image is calculated, whether the fiber uniformity belongs to a preset interval or not is judged, if not, the fiber putting speed, tension, temperature and humidity of the module area are input into a prediction model, and the fiber prediction uniformity in the specified time in the future is obtained; judging whether the fiber prediction uniformity belongs to a preset interval or not; and adjusting the fiber feeding speed and tension of the module area based on the analysis processing result. According to the method, the change trend of the fiber uniformity can be quickly identified and predicted, potential parameter anomalies can be identified in advance, production management personnel can take preventive measures before problems occur, meanwhile, the calculation amount can be effectively reduced, adjustment measures can be implemented more quickly, and therefore the production efficiency and the product quality are improved.

Description

Technical Field

[0001] The present application relates to the technical field of nonwoven fabric production monitoring, and in particular to an intelligent monitoring method and system for a high-performance fiber nonwoven fabric production line. Background Art

[0002] The production process of high-performance non-woven fabrics includes steps such as fiber pretreatment, fiber web forming, pre-curing, hot air setting and winding. First, the fibers are pre-treated and mixed, and then the fiber web is formed by spunbonding, meltblowing, etc., followed by pre-curing to enhance structural stability. Subsequently, the fiber web is formed into a stable structure through hot air setting, and finally wound into a roll to complete the production of high-performance non-woven fabrics.

[0003] At present, the production of high-performance non-woven fabrics usually monitors the parameters on the production line through sensors and takes adjustment measures when the parameters are abnormal. Its responsiveness is too dependent on the occurrence of abnormal events. The system takes adjustment measures after the parameter abnormality has occurred. The detection and resolution of potential problems in the production process are relatively slow. This delay may lead to production interruptions, unstable quality and waste of raw materials, causing significant losses to production efficiency and economic benefits. Summary of the invention

[0004] The present application provides an intelligent monitoring method and system for a high-performance fiber non-woven fabric production line, thereby solving the problem in the prior art that high-performance non-woven fabric production usually monitors parameters on the production line through sensors, and taking adjustment measures when the parameters are abnormal, which easily leads to production interruptions, unstable quality and waste of raw materials. Adjustment measures can be implemented more quickly, thereby improving production efficiency and product quality.

[0005] In a first aspect, the present application provides an intelligent monitoring method for a high-performance fiber non-woven fabric production line, which includes the following steps: obtaining a data set of the non-woven fabric production line, the data set including: dividing the fibers into a number of module areas along the width direction of the fibers, using a fast sensor to monitor the fiber delivery speed of each module area, the tension of the fibers before and after entering the web laying machine, the temperature, and the humidity, and obtaining a fiber image of each module area; calculating the fiber uniformity in the fiber image, and determining whether the fiber uniformity belongs to a preset interval, if not, inputting the fiber delivery speed, tension, temperature, and humidity of the module area into a prediction model to obtain a predicted fiber uniformity at a specified time in the future; determining whether the predicted fiber uniformity belongs to a preset interval, and if so, using a fusion data algorithm to perform data analysis and processing on the real-time fiber delivery speed, tension, temperature, and humidity to obtain an analysis and processing result; and adjusting the fiber delivery speed and tension of the module area based on the analysis and processing result.

[0006] Furthermore, obtaining the fiber image of each module area specifically includes: photographing the fiber image of the module area once at a specified interval, and the specified interval t is: Where S represents the length of the fiber in the fiber image, and v represents the movement speed of the fiber.

[0007] Furthermore, calculating the fiber uniformity in the fiber image specifically includes: using image processing technology to separate the fiber from the background in the fiber image to obtain a segmented image containing the fiber area and the background area; dividing the fiber area into N independent sub-areas, and obtaining the pixel density of the sub-areas, the average pixel density of all sub-areas and the number of sub-areas; calculating the fiber uniformity of the fiber area, and the fiber uniformity U is: Where D i represents the pixel density of the ith fiber region, Represents the average pixel density of all sub-regions in this fiber region.

[0008] Further, N is: N=k*total_pixels*min(width,height); wherein k represents a constant used to adjust the ratio of N, total_pixels represents the total number of pixels in the fiber image, min(width,height) represents the smaller value of the width or length of the fiber image, and min(width,height)≥1.

[0009] Furthermore, a fusion data algorithm is used to perform data analysis and processing on real-time tension, temperature and humidity, including: performing data fusion on real-time fiber delivery speed, tension, temperature and humidity to obtain fused data; performing anomaly detection on the fused data to obtain anomaly points; and performing trend analysis near the anomaly points.

[0010] Furthermore, the weighted average method is used to fuse the fiber delivery speed, tension, temperature and humidity data.

[0011] Furthermore, a trend analysis is performed near the abnormal point, including: calculating the correlation between the fiber delivery speed, tension, temperature and humidity at the abnormal point and the data at other time points, and obtaining the result of the correlation analysis.

[0012] Furthermore, adjusting the fiber delivery speed and tension of the module area based on the analysis and processing results specifically includes: determining whether the change in tension is the main factor causing poor fiber uniformity, and if so, adjusting the tension and monitoring the fiber uniformity after the adjustment until the tension reaches a preset limit or the fiber uniformity is within a preset range; if the tension reaches the preset limit, adjusting the fiber delivery speed and monitoring the fiber uniformity until the fiber delivery speed reaches the preset limit or the fiber uniformity is within a preset range; if the fiber delivery speed reaches the preset limit, sending an early warning message to the management personnel.

[0013] In the second aspect, the present application provides an intelligent monitoring system for a high-performance fiber non-woven fabric production line, which adopts the intelligent monitoring of the high-performance fiber non-woven fabric production line as in the first aspect, and is characterized in that it includes: a data acquisition module, a fiber prediction module, an analysis and processing module, and a production adjustment module.

[0014] The data acquisition module is used to obtain the data set of the non-woven fabric production line. The data set includes: dividing the fibers into several module areas along the width direction of the fibers, using fast sensors to monitor the fiber delivery speed of each module area, the tension, temperature, and humidity of the fibers before and after entering the web laying machine, and obtaining the fiber image of each module area.

[0015] The fiber prediction module is used to calculate the fiber uniformity in the fiber image and determine whether the fiber uniformity belongs to a preset interval. If so, the fiber delivery speed, tension, temperature and humidity of the module area are input into the trained prediction model to obtain the predicted fiber uniformity at a specified time in the future.

[0016] The analysis and processing module is used to determine whether the predicted fiber uniformity belongs to the preset range. If so, a fusion data algorithm is used to analyze and process the real-time fiber delivery speed, tension, temperature and humidity to obtain the analysis and processing results.

[0017] The production adjustment module is used to adjust the fiber delivery speed and tension of the module area based on the analysis and processing results.

[0018] In a third aspect, the present application provides an intelligent monitoring device for a high-performance fiber nonwoven fabric production line, comprising a memory and a processor.

[0019] The memory is used to store computer programs; the processor is used to implement the steps of the intelligent monitoring method for the high-performance fiber nonwoven fabric production line of the first aspect when executing the computer program.

[0020] The technical solution provided by this application has at least the following technical effects or advantages:

[0021] The intelligent monitoring and automatic adjustment mechanism is adopted to solve the problem that in the existing technology, high-performance non-woven fabric production usually monitors the parameters on the production line through sensors, and taking adjustment measures when the parameters are abnormal can easily lead to production interruption, unstable quality and waste of raw materials. It realizes real-time monitoring and regulation of fiber uniformity, can quickly identify and predict the changing trend of fiber uniformity, identify potential parameter abnormalities in advance, and enable production management personnel to take preventive measures before problems occur, avoid production interruption and quality problems, thereby reducing production losses. When the predicted fiber uniformity does not fall within the preset range, it can effectively reduce the amount of calculation and implement adjustment measures more quickly, thereby improving production efficiency and product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a flow chart of the intelligent monitoring method for the high-performance fiber nonwoven fabric production line in this application;

[0023] Figure 2 This is a flow chart for calculating fiber uniformity in a fiber image in this application. DETAILED DESCRIPTION

[0024] This application obtains the data set of the non-woven fabric production line in real time, calculates the fiber uniformity, and processes and analyzes the real-time data in combination with the fusion data algorithm. It can not only accurately predict the fiber uniformity, but also adjust the fiber delivery speed and tension in real time, thereby improving production efficiency and product quality.

[0025] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0026] Embodiment 1:

[0027] An intelligent monitoring method for a high-performance fiber nonwoven fabric production line comprises the following steps:

[0028] S100. Obtain a data set of a non-woven fabric production line, the data set including: dividing the fibers into a plurality of module areas along the width direction of the fibers, using fast sensors to monitor the fiber delivery speed of each module area, the tension of the fibers before and after entering the web laying machine, the temperature, and the humidity, and obtaining a fiber image of each module area.

[0029] S200. Calculate the fiber uniformity in the fiber image and determine whether the fiber uniformity belongs to a preset range. If not, input the fiber delivery speed, tension, temperature and humidity of the module area into the prediction model to obtain the predicted fiber uniformity at a specified time in the future.

[0030] S300. Determine whether the predicted fiber uniformity is within a preset range. If so, use a fusion data algorithm to analyze and process the real-time fiber delivery speed, tension, temperature and humidity to obtain analysis and processing results.

[0031] S400. Adjust the fiber delivery speed and tension of the module area based on the analysis and processing results.

[0032] This application uses an intelligent monitoring system to obtain data on the production line in real time, and uses image processing and machine learning technology to judge and predict the uniformity of the fiber. The relevant data of the fiber is obtained through a fast sensor, and the uniformity of the fiber is calculated through image processing. If the uniformity is abnormal, the system will use the trained prediction model to estimate the future fiber uniformity, and then determine whether data fusion and analysis are needed, and automatically adjust the production parameters according to the analysis and processing results to ensure that the uniformity of the fiber remains within the preset range, thereby improving the production quality of non-woven fabrics.

[0033] The intelligent monitoring and automatic adjustment mechanism is adopted to realize real-time monitoring and regulation of fiber uniformity, which can quickly identify and predict the changing trend of fiber uniformity and identify potential parameter abnormalities in advance, so that production managers can take preventive measures before problems occur, avoid production interruptions and quality problems, and thus reduce production losses. When the predicted fiber uniformity does not fall within the preset range, it can effectively reduce the amount of calculation and implement adjustment measures more quickly, thereby improving production efficiency and product quality.

[0034] The acquisition of data sets can be based on different types of sensors, the prediction model can be a deep learning algorithm, and the judgment criteria and preset intervals of fiber uniformity can be adjusted according to the specific requirements of the product to meet the production of non-woven fabrics of different specifications and uses.

[0035] In S100, the fiber image of each module area is obtained by taking a fiber image of the module area once at a specified interval, and the specified interval t is: Where S represents the length of the fiber in the fiber image, and v represents the movement speed of the fiber.

[0036] By determining the time interval for taking images based on the length and movement speed of the fiber, image data can be evenly acquired during the fiber movement process, ensuring the coverage of the image to the fiber and ensuring that the acquisition of the fiber image can accurately reflect the real-time status of the fiber on the production line, thereby improving the accuracy of subsequent fiber uniformity calculation and analysis.

[0037] By dynamically adjusting the interval between image capture, the data deviation caused by changes in fiber movement speed during image acquisition can be effectively reduced, ensuring the representativeness of image data, improving the monitoring accuracy of fiber uniformity, and enhancing the reliability and accuracy of the prediction model, thereby further optimizing the monitoring efficiency of the production line and product quality.

[0038] In S200, calculating the fiber uniformity in the fiber image specifically includes:

[0039] S210. Use image processing technology to separate the fibers in the fiber image from the background, and obtain a segmented image including a fiber area and a background area.

[0040] S210. Divide the fiber region into N independent sub-regions, and obtain the pixel density of the sub-region, the average pixel density of all sub-regions, and the number of sub-regions.

[0041] S230. Calculate the fiber uniformity of the fiber area, the fiber uniformity U is: Where D i represents the pixel density of the ith fiber region, Represents the average pixel density of all sub-regions in this fiber region.

[0043] Image processing technology is used to separate the fibers from the background in the fiber image so that a more detailed analysis of the fiber area can be performed. By dividing the fiber area into multiple sub-areas and calculating the pixel density of each sub-area, the uniformity of the fibers can be effectively quantified. The calculation formula of the uniformity U takes into account the deviation of the pixel density of each sub-area relative to the average value, which can accurately reflect the uniform distribution of the fibers.

[0044] Through refined image segmentation and region division, the fiber distribution can be analyzed more comprehensively. This method can effectively detect small unevenness in fiber distribution, which helps to adjust production parameters in time, thereby improving the quality of nonwoven products.

[0045] Convolutional neural networks (CNNs) can be used for image segmentation to improve the accuracy of fiber and background separation. In addition, the calculation formula of fiber uniformity U can also be adjusted according to different application scenarios, using different statistical methods or increasing weight coefficients to adapt to special fiber distribution characteristics.

[0046] In S210, N is: N = k*total_pixels*min(width, height); wherein k represents a constant used to adjust the ratio of N, total_pixels represents the total number of pixels in the fiber image, min(width, height) represents the smaller value of the width or length of the fiber image, and min(width, height) ≥ 1.

[0047] The constant k is set to flexibly adjust the division of the fiber area in the image, ensuring that the number of sub-areas N is related to the resolution and size of the fiber image. The fiber area can be divided into an appropriate number of sub-areas by calculating the number of sub-areas N, thereby calculating the pixel density of each sub-area, ensuring the accuracy of the fiber uniformity calculation and avoiding the influence of too many or too few sub-areas on the analysis results.

[0048] The adjustment of the proportional constant k can better adapt to fiber images of different sizes and resolutions, maintain the flexibility of the algorithm under different production line conditions, and thus improve the adaptability and accuracy of the intelligent monitoring system for nonwoven production lines. The number of sub-regions N can also be modified according to other image characteristics or monitoring objectives. For example, considering the non-uniformity of fiber density distribution, different segmentation algorithms can be used to determine the division method of sub-regions, thereby improving the robustness and applicability of the monitoring system.

[0049] In S300, the fusion data algorithm is used to analyze and process the real-time tension, temperature and humidity data, including:

[0050] S310. Perform data fusion on the real-time fiber delivery speed, tension, temperature and humidity to obtain fused data.

[0051] The real-time fiber delivery speed, tension, temperature and humidity are fused to obtain fused data. The data fusion adopts the weighted average method to ensure that the relative importance of each parameter is properly reflected in the fusion process.

[0052] S320. Perform anomaly detection on the fused data to obtain anomaly points.

[0053] Anomaly detection is performed on the fused data to identify abnormal points. Anomaly detection can be achieved by setting thresholds or using machine learning models to accurately identify abnormal situations that may affect fiber uniformity.

[0054] S330. Perform trend analysis near the outlier point.

[0055] Perform trend analysis near the abnormal point and evaluate the data change trend before and after the abnormal point to determine whether there is a persistent process deviation or a short-term abnormal fluctuation.

[0056] By integrating multiple real-time monitoring parameters into a comprehensive data set, the subsequent analysis steps are simplified. The weighted average method is used to fuse the fiber delivery speed, tension, temperature and humidity data, which can balance the influence of each parameter during the fusion process. Anomaly detection and trend analysis further ensure that the system can identify and respond to potential problems in a timely manner, so as to make rapid adjustments when the fiber uniformity begins to deviate significantly from the standard.

[0057] This implementation method improves the accuracy and response speed of data analysis, and can more timely discover and handle abnormal situations in the production process. Through data fusion technology, the system can better balance the impact of various parameters and reduce the interference of data fluctuations on monitoring results, thereby enhancing the overall stability and product quality of the non-woven production line.

[0058] Data fusion methods can choose other algorithms, such as Bayesian estimation or neural network models, to adapt to more complex process conditions. Anomaly detection and trend analysis methods can also be adjusted according to actual needs, such as using more complex time series analysis methods or introducing more data sources to improve monitoring accuracy. In addition, the weight distribution during data fusion can be adjusted according to the characteristics of different production lines to better adapt to actual production conditions.

[0059] In S330, a trend analysis is performed near the abnormal point, including: calculating the correlation between the fiber delivery speed, tension, temperature and humidity at the abnormal point and the data at other time points, and obtaining the result of the correlation analysis.

[0060] By analyzing the production parameters at the abnormal point and its vicinity and identifying the correlation between these parameters, potential causal relationships can be revealed. For example, tension changes may affect fiber uniformity, which can more accurately locate the source of the problem and guide production adjustments. By revealing the complex relationship between multiple variables, it helps identify the key factors that affect fiber uniformity, improves the monitoring and adjustment accuracy of the production line, and ultimately improves product quality and production efficiency.

[0061] In S400, adjusting the fiber delivery speed and tension of the module area based on the analysis and processing results specifically includes:

[0062] S410. Determine whether the change in tension is the main factor causing poor fiber uniformity. If so, adjust the tension and monitor the fiber uniformity after the adjustment until the tension reaches a preset limit or the fiber uniformity is within a preset range.

[0063] S420. If the tension reaches the preset limit, the fiber delivery speed is adjusted and the fiber uniformity is monitored until the fiber delivery speed reaches the preset limit or the fiber uniformity is within the preset range.

[0064] S430. If the fiber delivery speed reaches a preset limit, a warning message is sent to the management personnel.

[0065] The fiber delivery speed and tension are adjusted in steps to optimize fiber uniformity. The tension is adjusted first. If the tension adjustment fails to effectively improve the uniformity, the fiber delivery speed is further adjusted until the control limit of the system is reached. When the system cannot restore the uniformity through automatic adjustment, an early warning message is sent to remind the management personnel to intervene manually.

[0066] Through the automated adjustment and monitoring mechanism, the time and frequency of manual intervention are reduced, the stability and efficiency of the production line are improved, and the early warning function can promptly indicate potential problems to avoid production interruptions or quality losses.

[0067] Embodiment 2:

[0068] The present embodiment provides an intelligent monitoring system for a high-performance fiber non-woven fabric production line, which adopts the intelligent monitoring of the high-performance fiber non-woven fabric production line as in Example 1, and is characterized in that it includes: a data acquisition module, a fiber prediction module, an analysis and processing module, and a production adjustment module.

[0069] The data acquisition module is used to obtain the data set of the non-woven fabric production line. The data set includes: dividing the fibers into several module areas along the width direction of the fibers, using fast sensors to monitor the fiber delivery speed of each module area, the tension, temperature, and humidity of the fibers before and after entering the web laying machine, and obtaining the fiber image of each module area.

[0070] The fiber prediction module is used to calculate the fiber uniformity in the fiber image and determine whether the fiber uniformity belongs to the preset range. If so, the fiber delivery speed, tension, temperature and humidity of the module area are input into the trained prediction model to obtain the predicted fiber uniformity at a specified time in the future.

[0071] The analysis and processing module is used to determine whether the predicted fiber uniformity belongs to the preset range. If so, a fusion data algorithm is used to analyze and process the real-time fiber delivery speed, tension, temperature and humidity to obtain the analysis and processing results.

[0072] The production adjustment module is used to adjust the fiber delivery speed and tension of the module area based on the analysis and processing results.

[0073] This embodiment has all the advantages of the intelligent monitoring method for a high-performance fiber nonwoven fabric production line in the first embodiment, and can automatically implement all steps of the intelligent monitoring method for a high-performance fiber nonwoven fabric production line.

[0074] Embodiment three:

[0075] This embodiment provides an intelligent monitoring device for a high-performance fiber nonwoven fabric production line, including a memory and a processor.

[0076] The memory is used to store computer programs; the processor is used to implement the steps of the intelligent monitoring method for the high-performance fiber nonwoven fabric production line in Example 1 when executing the computer program.

[0077] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps of the functions specified in a box or multiple boxes. 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 know the basic creative concept. Therefore, the attached 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.

Claims

1. An intelligent monitoring method for a high-performance fiber nonwoven fabric production line, characterized in that: The following steps are involved: Obtain a data set of a non-woven fabric production line. The data set includes: dividing the fibers into several module areas along the width direction of the fibers, using fast sensors to monitor the fiber delivery speed of each module area, the tension, temperature, and humidity of the fibers before and after entering the web laying machine, and obtaining fiber images of each module area; Calculate the fiber uniformity in the fiber image and determine whether the fiber uniformity belongs to a preset interval. If not, input the fiber delivery speed, tension, temperature and humidity of the module area into the prediction model to obtain the predicted fiber uniformity at a specified time in the future. Determine whether the predicted fiber uniformity is within the preset range. If so, use a fusion data algorithm to analyze and process the real-time fiber delivery speed, tension, temperature and humidity to obtain the analysis and processing results; The fiber delivery speed and tension of the module area are adjusted based on the analysis and processing results.

2. The intelligent monitoring method for a high-performance fiber nonwoven fabric production line according to claim 1, characterized in that: The fiber image of each module area is obtained by taking a fiber image of the module area once at a specified interval, and the specified interval t is: Where S represents the length of the fiber in the fiber image, and v represents the movement speed of the fiber.

3. The intelligent monitoring method for a high-performance fiber nonwoven fabric production line according to claim 1, characterized in that: Calculating the fiber uniformity in the fiber image specifically includes: Using image processing technology to separate the fibers from the background in the fiber image, a segmented image containing the fiber area and the background area is obtained; Divide the fiber area into N independent sub-areas, and obtain the pixel density of the sub-areas, the average pixel density of all sub-areas, and the number of sub-areas; Calculate the fiber uniformity of the fiber area, the fiber uniformity U is: Where D i represents the pixel density of the ith fiber region, Represents the average pixel density of all sub-regions in this fiber region.

4. The intelligent monitoring method for a high-performance fiber nonwoven fabric production line according to claim 3, characterized in that: N is: N=k*total_pixels*min(width,height) Wherein, k represents a constant used to adjust the ratio of N, total_pixels represents the total number of pixels in the fiber image, min(width,height) represents the smaller value of the fiber image width or length, and min(width,height)≥1.

5. The intelligent monitoring method for a high-performance fiber nonwoven fabric production line according to claim 1, characterized in that: The fusion data algorithm is used to analyze and process the real-time tension, temperature and humidity data, including: Perform data fusion on real-time fiber delivery speed, tension, temperature and humidity to obtain fused data; Perform anomaly detection on the fused data to obtain abnormal points; Perform trend analysis around outliers.

6. The intelligent monitoring method for a high-performance fiber nonwoven fabric production line according to claim 5, characterized in that: The weighted average method is used to fuse the data of fiber delivery speed, tension, temperature and humidity.

7. The intelligent monitoring method for a high-performance fiber nonwoven fabric production line according to claim 5, characterized in that: Trend analysis is performed near the abnormal point, including: calculating the correlation between the fiber delivery speed, tension, temperature and humidity at the abnormal point and the data at other time points, and obtaining the result of the correlation analysis.

8. The intelligent monitoring method for a high-performance fiber nonwoven fabric production line according to claim 1, characterized in that: Adjusting the fiber delivery speed and tension of the module area based on the analysis and processing results specifically includes: Determine whether the change in tension is the main factor causing poor fiber uniformity. If so, adjust the tension and monitor the fiber uniformity after the adjustment until the tension reaches a preset limit or the fiber uniformity is within a preset range; If the tension reaches the preset limit, the fiber delivery speed is adjusted and the fiber uniformity is monitored until the fiber delivery speed reaches the preset limit or the fiber uniformity is within the preset range; If the fiber delivery rate reaches the preset limit, an early warning message will be sent to the management personnel.

9. An intelligent monitoring system for a high-performance fiber nonwoven production line, characterized in that: include: Data acquisition module: It is used to obtain the data set of the non-woven fabric production line. The data set includes: dividing the fibers into several module areas along the width direction of the fibers, using fast sensors to monitor the fiber delivery speed of each module area, the tension, temperature, and humidity of the fibers before and after entering the web laying machine, and obtaining the fiber image of each module area; Fiber prediction module: It is used to calculate the fiber uniformity in the fiber image and determine whether the fiber uniformity belongs to the preset interval. If so, the fiber delivery speed, tension, temperature and humidity of the module area are input into the trained prediction model to obtain the predicted fiber uniformity at a specified time in the future; Analysis and processing module: It is used to determine whether the predicted fiber uniformity falls within the preset range. If so, a fusion data algorithm is used to analyze and process the real-time fiber delivery speed, tension, temperature and humidity to obtain the analysis and processing results; Production adjustment module: It is used to adjust the fiber delivery speed and tension of the module area based on the analysis and processing results.

Images