High-speed flexible base material production monitoring system based on data analysis

Abstract

The invention relates to the field of base material production monitoring, in particular to a high-speed flexible base material production monitoring system based on data analysis. Through cooperative work of the acquisition module, the pre-analysis module, the verification module and the intervention adjustment module, the acquisition module acquires the tension, the spatial distance and the surface image data of the flexible substrate, and the pre-analysis module determines the tension variation interval by calculating the tension difference degree. A base material variation region is further determined by calculating a space distance difference value, a verification module verifies the continuity of the tension variation region, a region distance distribution diagram is constructed according to a response and verification result, and a deviation trend is judged. And the intervention adjustment module adjusts the base material production parameters in the time domain dimension gradient according to the deviation trend, and determines the optimal base material production parameters based on the deviation value, thereby improving the real-time monitoring and accurate regulation capability in the base material production process, and further improving the reliability of early monitoring and the stability of the production process.

Description

Technical Field

[0001] This invention relates to the field of substrate production monitoring, and more particularly to a high-speed flexible substrate production monitoring system based on data analysis. Background Technology

[0002] With the booming development of strategic emerging industries such as flexible electronics, new energy batteries, and advanced packaging, the production quality of high-speed flexible substrates, as their core carriers and key components, is not only closely related to the performance, reliability, and service life of end products, but also closely linked to the efficiency, cost, and material utilization of the production process. In this high-speed, continuous roll-to-roll manufacturing process, the high coupling of process parameters and the instantaneous nature of production conditions mean that even minute steady-state deviations generated in any process can be rapidly transmitted and amplified under high-speed stretching, easily leading to continuous defective products hundreds of meters long in a short period, causing significant economic losses. Therefore, real-time, accurate, and intelligent online monitoring and analysis, and predictive control based on this, have become an indispensable part of breaking through the bottlenecks in yield and efficiency in high-end flexible substrate manufacturing and promoting the industry's digital and intelligent upgrade.

[0003] Chinese Patent Publication No. CN120525952A relates to the field of visual inspection technology for coating defects, specifically to a method for detecting coating defects on the surface of plastic substrates based on LMC technology. The method includes: acquiring a grayscale image of the coating morphology on the surface of the plastic substrate; constructing the pit direction difference of each pixel; obtaining the pit edge crack feature value of each pixel; further obtaining the pit feature value of each pixel; extracting suspected pit feature points; combining the positional relationship between each pixel and the suspected pit feature points to obtain the pit density saliency of each pixel; using a quaternary Fourier saliency detection algorithm to obtain a saliency map of the coating pits; and obtaining the region of coating pit defects through image segmentation. This application can improve the detection accuracy of coating defects.

[0004] However, the following problems still exist in the existing technology.

[0005] In the continuous roll-to-roll production of high-speed flexible substrates, due to the high coupling and instantaneous nature of the production process, any small steady-state deviation generated in any process will be rapidly transmitted and replicated under high-speed stretching. This may cause defects to be extended in the longitudinal direction and solidified in the transverse direction, thereby producing continuous defective products up to hundreds of meters long in a very short time, thus reducing the overall yield of the whole roll product and the overall operating efficiency of the production line. Summary of the Invention

[0006] To address this, the present invention provides a high-speed flexible substrate production monitoring system based on data analysis, which overcomes the problem in the prior art where, due to the highly coupled and instantaneous nature of the production process, any small steady-state deviation generated in any process can be rapidly transmitted and replicated under high-speed stretching. This may cause defects to be extended longitudinally and solidified laterally, resulting in the production of continuous defective products up to hundreds of meters long in a very short time, thereby reducing the overall yield of the whole roll of products and the overall operating efficiency of the production line.

[0007] To achieve the above objectives, the present invention provides a high-speed flexible substrate production monitoring system based on data analysis, comprising:

[0008] The acquisition module includes a tension detection roller for acquiring tension in various regions of the flexible substrate, a laser ranging unit for detecting the spatial distance corresponding to the relative fixed height of each region, and an image acquisition unit for acquiring images of the surface of the flexible substrate.

[0009] The pre-analysis module, which is connected to the acquisition module, is used to record the tension in each region in real time to determine the tension difference, determine the tension variation range based on the tension difference in each time domain segment, and determine the substrate variation region based on the difference between the relative distance thresholds of the corresponding spatial distances in each region.

[0010] The verification module, which is connected to the pre-analysis module, is used to perform time variation verification based on the continuity of the tension variation range. In response to the result of the time variation verification, the spatial distance of each detected area within the tension variation range is extracted, a regional distance distribution map is constructed, and the deviation trend is determined based on the proportion of the substrate variation area in the regional distance distribution map.

[0011] An intervention and adjustment module, which is connected to the acquisition module, the pre-analysis module, and the verification module respectively, adjusts the substrate production parameters in the time domain dimension in response to the deviation trend, acquires the surface images of the flexible substrate corresponding to each time domain segment, determines the deviation of the surface images of the flexible substrate, determines the reference time domain segment, and determines the optimal substrate production parameters based on the reference time domain segment.

[0012] The substrate production parameters include the tension setting value and the conveying speed of the flexible substrate.

[0013] Furthermore, the pre-analysis module is characterized in that it records the tension in each region in real time to determine the tension difference, including:

[0014] Used to determine the tension variance of each region detected at each time point within the time domain segment, and to determine the mean of the tension variance;

[0015] This is used to determine the mean of the tension variance as the tension difference degree.

[0016] Furthermore, the pre-analysis module is characterized in that it determines the tension variation interval based on the tension difference within each time domain segment, including:

[0017] If the tension difference within the time domain segment is greater than or equal to a predetermined tension difference threshold, then the time domain segment is determined to be a tension variation interval.

[0018] Further, the pre-analysis module is characterized in that it determines the substrate variation region based on the difference between the relative distance thresholds of the corresponding spatial distances of each region, including:

[0019] Used to determine the difference between the relative distance thresholds corresponding to the spatial distances of each region;

[0020] If the difference is greater than or equal to a preset difference threshold, the region is determined as a substrate variation region.

[0021] Furthermore, the verification module is characterized in that it performs time-varying verification based on the continuity of the tension variation range, including:

[0022] This is used to determine whether the tension variation range meets the continuity condition. If the continuity condition is met, the time variation verification is passed.

[0023] The continuity condition is that the tension variation intervals are continuous and the number of consecutive tension variation intervals is greater than a predetermined threshold.

[0024] Furthermore, the verification module is characterized in that the results of the response time variation verification include,

[0025] If the tension variation range passes the time variation verification, the spatial distance of each detected area within the tension variation range is extracted, and a regional distance distribution map is constructed. The deviation trend is determined based on the proportion of the substrate variation area in the regional distance distribution map.

[0026] Further, the verification module is characterized in that it extracts the spatial distance of each detected region within the tension variation range and constructs a region distance distribution map, including:

[0027] Used to determine the regional layout map corresponding to each region at a single moment;

[0028] This is used to arrange the regional layout map according to the time sequence to obtain the regional distance distribution map;

[0029] Used to mark substrate variation areas based on the regional distance distribution map.

[0030] Furthermore, the verification module is characterized in that it determines the deviation trend based on the proportion of substrate variation areas in the regional distance distribution map, including:

[0031] Used to determine the proportion of substrate variation areas in the regional distance distribution map;

[0032] If the percentage is greater than or equal to a predetermined percentage threshold, it is determined that the flexible substrate has a deviation trend.

[0033] Further, the feature is that the intervention adjustment module is used to determine the deviation amount of the flexible substrate surface image, and determining the reference time domain segment includes,

[0034] Used to determine the image similarity between the flexible substrate surface image and the corresponding standard image for each time domain segment;

[0035] The reciprocal of the image similarity is used to determine the deviation.

[0036] This is used to determine the time domain segment corresponding to the deviation as the reference time domain segment;

[0037] Among them, adjusting the substrate production parameters in the time domain dimension includes changing the substrate production parameters at predetermined intervals.

[0038] Further, the feature is that the intervention adjustment module, used to determine the optimal substrate production parameters based on the reference time domain segment, includes:

[0039] The substrate production parameters, which are gradient-adjusted within the reference time domain, are used to determine the optimal substrate production parameters.

[0040] Compared with existing technologies, this invention utilizes a collaborative approach involving a data acquisition module, a pre-analysis module, a verification module, and an intervention / adjustment module. The acquisition module obtains tension, spatial distance, and surface image data of the flexible substrate. The pre-analysis module determines the tension variation range by calculating tension differences and further identifies the substrate variation region by calculating spatial distance differences. The verification module verifies the continuity of the tension variation range, constructing a regional distance distribution map based on the response and verification results to determine the deviation trend. The intervention / adjustment module then adjusts the substrate production parameters in a gradient manner across the time domain according to the deviation trend, determining the optimal substrate production parameters based on the deviation amount. This enhances the real-time monitoring and precise control capabilities during substrate production, thereby improving the reliability of early monitoring and the stability of the production process.

[0041] In particular, this invention determines the tension difference by recording the tension in each region in real time, determines the tension variation range based on the tension difference within each time domain segment, and determines the substrate variation region based on the difference between the relative distance thresholds of the corresponding spatial distances in each region. In continuous roll-to-roll production of high-speed flexible substrates, due to the high coupling and instantaneous nature of the production process, any small steady-state deviation generated in any process will be rapidly transmitted and replicated under high-speed drawing, which may cause defects to be extended longitudinally and solidified laterally. Since tension fluctuations directly affect the forming quality and physical properties of the substrate, and small changes in tension may be rapidly amplified during high-speed production, leading to a decrease in the mechanical properties of the substrate and surface quality defects, determining the tension variation range based on the tension difference within each time domain segment can more accurately identify tension fluctuations that may occur during the production process. Since changes in the thickness or flatness of the substrate can affect its performance and quality, and changes in spatial distance can characterize changes in the thickness or flatness of the substrate, the laser ranging unit can accurately measure the distance between the substrate surface and the distance threshold. For example, when the substrate becomes thinner, the spatial distance detected by the laser ranging unit will increase. Therefore, the substrate variation area is determined based on the difference between the spatial distance of each region and the distance threshold. Through these measures, a reliable basis can be provided for the subsequent production process, improving the real-time monitoring and precise control capabilities in the substrate production process, thereby improving the reliability of early monitoring and the stability of the production process.

[0042] In particular, this invention verifies temporal variation based on the continuity of tension variation intervals. Responding to the results of this temporal variation verification, it extracts the spatial distances of each detected region within the tension variation interval, constructs a region distance distribution map, and determines the deviation trend based on the proportion of substrate variation regions in the region distance distribution map. In continuous roll-to-roll production of high-speed flexible substrates, the identified individual "tension variation interval" signals may be sporadic and isolated, and do not represent actual process instability. For example, external vibrations of the equipment may cause momentary tension anomalies. Directly triggering system adjustments based on this would lead to unnecessary malfunctions and interfere with stable production. Therefore, temporal variation verification based on the continuity of tension variation intervals is used to more accurately filter out random, instantaneous interference. However, simply confirming the continued existence of anomalies through time verification is insufficient to guide precise control. Because a persistent anomaly requires quantitative assessment of its scope and severity, for example, if the anomaly only occurs in a fixed, narrow area at the edge of the substrate width, its cause is more likely to stem from local mechanical components, such as the periodic effects of minor scratches on the guide roller edge or bearing defects. This usually has limited impact and may not require immediate adjustment of global process parameters. Therefore, constructing a regional distance distribution map and calculating the proportion of the substrate's anomalous area reflects the degree of problem diffusion across the substrate's transverse width during the verified anomaly period. A high proportion of anomaly areas indicates that the problem is not localized or isolated, but rather widely distributed or occurring in clusters. This suggests a potential overall, system-level parameter imbalance or state deviation in the production system, posing a risk of disrupting the continuity of the entire roll. Early intervention and adjustment enhance the real-time monitoring and precise control capabilities during substrate production, thereby improving the reliability of early monitoring and the stability of the production process.

[0043] In particular, this invention adjusts the substrate production parameters in the time domain dimension by gradient adjustment, collects flexible substrate surface images corresponding to each time domain segment, determines the deviation of the flexible substrate surface images, determines a reference time domain segment, and determines the optimal substrate production parameters based on the reference time domain segment. In the closed-loop control process of high-speed flexible substrates, due to the complex dynamic characteristics of the production system, a one-time, large-amplitude parameter adjustment can easily cause oscillations. Therefore, a gradient adjustment strategy in the time domain dimension is adopted. The core of this strategy is to transform the parameter adjustment action into an orderly, controlled, and observable system optimization process. Each adjustment step simultaneously collects the substrate surface image produced under the action of these parameters and calculates its deviation from the standard image to determine the reference time domain segment. The corresponding parameters are then determined as the optimal substrate production parameters. The aim is to transform experience-dependent, black-box parameter adjustment into an optimization based on online quality feedback, reducing production instability caused by parameter mutations, improving real-time monitoring and precise control capabilities in the substrate production process, and thus enhancing the reliability of early monitoring and the stability of the production process. Attached Figure Description

[0044] Figure 1 This is a system block diagram of a high-speed flexible substrate production monitoring system based on data analysis, as an embodiment of the invention.

[0045] Figure 2 This is a logic diagram for determining the substrate variation region in an embodiment of the invention;

[0046] Figure 3 This is a schematic diagram of the regional distance distribution map according to an embodiment of the invention;

[0047] Figure 4 This is a logic diagram for determining the deviation trend in an embodiment of the invention.

[0048] 1: Substrate variation area. Detailed Implementation

[0049] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0050] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0051] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the term "connection" should be interpreted broadly. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0052] Please see Figure 1 The diagram shown is a system block diagram of a high-speed flexible substrate production monitoring system based on data analysis, according to an embodiment of the invention. The system includes:

[0053] The acquisition module includes a tension detection roller for acquiring tension in various regions of the flexible substrate, a laser ranging unit for detecting the spatial distance corresponding to the relative fixed height of each region, and an image acquisition unit for acquiring images of the surface of the flexible substrate.

[0054] The pre-analysis module, which is connected to the acquisition module, is used to record the tension in each region in real time to determine the tension difference, determine the tension variation range based on the tension difference in each time domain segment, and determine the substrate variation region based on the difference between the relative distance thresholds of the corresponding spatial distances in each region.

[0055] The verification module, which is connected to the pre-analysis module, is used to perform time variation verification based on the continuity of the tension variation range. In response to the result of the time variation verification, the spatial distance of each detected area within the tension variation range is extracted, a regional distance distribution map is constructed, and the deviation trend is determined based on the proportion of the substrate variation area in the regional distance distribution map.

[0056] An intervention and adjustment module, which is connected to the acquisition module, the pre-analysis module, and the verification module respectively, adjusts the substrate production parameters in the time domain dimension in response to the deviation trend, acquires the surface images of the flexible substrate corresponding to each time domain segment, determines the deviation of the surface images of the flexible substrate, determines the reference time domain segment, and determines the optimal substrate production parameters based on the reference time domain segment.

[0057] The substrate production parameters include the tension setting value and the conveying speed of the flexible substrate.

[0058] Specifically, there are no restrictions on the structure of the pre-analysis module, the verification module, and the intervention and adjustment module. They can be composed of logic components or combinations of logic components, including field-programmable processors, computers, or microprocessors in computers.

[0059] Specifically, there are no restrictions on the structure of the acquisition module. The laser ranging unit can use a high-precision non-contact laser sensor or integrate a multi-line lidar system to achieve accurate measurement of the spatial distance corresponding to the relatively fixed height of each area of ​​the flexible substrate. The image acquisition unit can use a high-resolution image sensor or utilize a high-speed monochrome camera, as long as it can accurately acquire the tension, spatial distance and surface image data of the flexible substrate.

[0060] The tension detection roller can be the tension detection roller disclosed in Chinese patent application publication number CN105841870A, to detect the tension in various areas of the flexible substrate, which will not be elaborated further here.

[0061] In implementation, laser ranging units are deployed at designated workstations on the production line perpendicular to the travel direction of the flexible substrate to fully cover the substrate. To balance monitoring resolution, system cost, and data processing efficiency, the preferred number of units is six.

[0062] Specifically, the pre-analysis module is used to record the tension in each region in real time to determine the tension difference, including:

[0063] Used to determine the tension variance of each region detected at each time point within the time domain segment, and to determine the mean of the tension variance;

[0064] This is used to determine the mean of the tension variance as the tension difference degree.

[0065] In practice, the time domain segment is usually selected within the interval [0.5s, 2.5s], and is preferably 2s.

[0066] Specifically, the pre-analysis module is used to determine the tension variation range based on the tension difference within each time domain segment, including:

[0067] If the tension difference within the time domain segment is greater than or equal to a predetermined tension difference threshold, then the time domain segment is determined to be a tension variation interval.

[0068] In implementation, the purpose of the tension difference threshold is to distinguish between normal random fluctuations and abnormal instability trends in tension distribution during production. Those skilled in the art can perform statistical analysis based on tension data from historical stable production periods to determine the mean of the tension variance, thus representing the tension variation under normal conditions. To indicate abnormal tension changes, the tension difference threshold is set as a predetermined multiple of the mean. Typically, this predetermined multiple is selected within the range of [0.15, 1.35], and is preferably 1.25 in implementation.

[0069] This invention determines tension difference by recording tension in each region in real time, identifies tension variation ranges based on tension differences within each time domain, and determines substrate variation regions based on the difference in relative distance thresholds between corresponding spatial distances of each region. In continuous roll-to-roll production of high-speed flexible substrates, due to the highly coupled and instantaneous nature of the production process, minute steady-state deviations generated in any step are rapidly transmitted and replicated under high-speed drawing, which may cause defects to be extended longitudinally and solidified laterally. Since tension fluctuations directly affect the forming quality and physical properties of the substrate, and minute changes in tension may be rapidly amplified during high-speed production, leading to a decrease in the mechanical properties of the substrate and surface quality defects, determining tension variation ranges based on tension differences within each time domain allows for more accurate identification of tension fluctuations that may occur during the production process. Since changes in the thickness or flatness of the substrate can affect its performance and quality, and changes in spatial distance can characterize changes in the thickness or flatness of the substrate, the laser ranging unit can accurately measure the distance between the substrate surface and the distance threshold. For example, when the substrate becomes thinner, the spatial distance detected by the laser ranging unit will increase. Therefore, the substrate variation area is determined based on the difference between the spatial distance of each region and the distance threshold. Through these measures, a reliable basis can be provided for the subsequent production process, improving the real-time monitoring and precise control capabilities in the substrate production process, thereby improving the reliability of early monitoring and the stability of the production process.

[0070] Please see Figure 2 As shown, this is a logic diagram for determining the substrate mutation region according to an embodiment of the invention. Specifically, the pre-analysis module is used to determine the substrate mutation region based on the difference between the relative distance thresholds of the corresponding spatial distances of each region, including:

[0071] Used to determine the difference between the relative distance thresholds corresponding to the spatial distances of each region;

[0072] If the difference is greater than or equal to a preset difference threshold, the region is determined as a substrate variation region.

[0073] In implementation, the purpose of the difference threshold is to characterize the significance of spatial distance changes in the substrate during the production process. The difference threshold is predetermined. Those skilled in the art can determine the mean of the spatial distance difference by statistically analyzing spatial distance data under normal production conditions to represent the changes in substrate spatial distance under normal circumstances. To represent abnormal changes in substrate spatial distance, the difference threshold is set as the product of the mean and the difference accuracy coefficient. Typically, the difference accuracy coefficient is selected within the range of [0.85, 1.35], and is preferably 1.25 in implementation.

[0074] Specifically, the verification module is used to perform time-varying verification based on the continuity of the tension variation range, including:

[0075] This is used to determine whether the tension variation range meets the continuity condition. If the continuity condition is met, the time variation verification is passed.

[0076] The continuity condition is that the tension variation intervals are continuous and the number of consecutive tension variation intervals is greater than a predetermined threshold.

[0077] In practice, to ensure the continuity of the tension variation range, the number threshold should be no less than 3, and preferably 5.

[0078] Specifically, the results of the verification module for responding to time variation verification include:

[0079] If the tension variation range passes the time variation verification, the spatial distance of each detected area within the tension variation range is extracted, and a regional distance distribution map is constructed. The deviation trend is determined based on the proportion of the substrate variation area in the regional distance distribution map.

[0080] Please see Figure 3 As shown, this is a schematic diagram of the regional distance distribution map according to an embodiment of the invention. The verification module is used to extract the spatial distance of each detected region within the tension variation range, and to construct the regional distance distribution map, including...

[0081] Used to determine the regional layout map corresponding to each region at a single moment;

[0082] This is used to arrange the regional layout map according to the time sequence to obtain the regional distance distribution map;

[0083] Used to mark substrate variation areas based on the regional distance distribution map.

[0084] In practice, the area distance distribution map is recorded along the direction of substrate travel, i.e., perpendicular to the direction of substrate rolling. As the substrate rolls continuously, the corresponding spatial distance data of each area is continuously collected, and the area distance distribution map is updated in real time. The substrate variation areas are marked on the area distance distribution map.

[0085] Please see Figure 4 As shown, this is a logic diagram for determining the deviation trend in an embodiment of the invention. Specifically, the verification module determines the deviation trend based on the proportion of substrate variation areas in the regional distance distribution map, including:

[0086] Used to determine the proportion of substrate variation areas in the regional distance distribution map;

[0087] If the percentage is greater than or equal to a predetermined percentage threshold, it is determined that the flexible substrate has a deviation trend.

[0088] In practice, the percentage threshold is usually selected within the range of [20%, 55%], and is preferably 45%.

[0089] This invention verifies temporal variations based on the continuity of tension variation intervals. Responding to the results of this verification, it extracts the spatial distances of each detected region within the tension variation interval, constructs a region distance distribution map, and determines the deviation trend based on the proportion of substrate variation regions in the map. In continuous roll-to-roll production of high-speed flexible substrates, the identified individual "tension variation interval" signals may be sporadic and isolated, not representing actual process instability. For example, external vibrations of the equipment may cause momentary tension anomalies. Directly triggering system adjustments based on this would lead to unnecessary malfunctions and disrupt stable production. Therefore, temporal variation verification based on the continuity of tension variation intervals more accurately filters out random, instantaneous interference. However, simply confirming the persistence of anomalies through time verification is insufficient to guide precise control. Because a persistent anomaly requires quantitative assessment of its scope and severity, for example, if the anomaly only occurs in a fixed, narrow area at the edge of the substrate width, its cause is more likely to stem from local mechanical components, such as the periodic effects of minor scratches on the guide roller edge or bearing defects. This usually has limited impact and may not require immediate adjustment of global process parameters. Therefore, constructing a regional distance distribution map and calculating the proportion of the substrate's anomalous area reflects the degree of problem diffusion across the substrate's transverse width during the verified anomaly period. A high proportion of anomaly areas indicates that the problem is not localized or isolated, but rather widely distributed or occurring in clusters. This suggests a potential overall, system-level parameter imbalance or state deviation in the production system, posing a risk of disrupting the continuity of the entire roll. Early intervention and adjustment enhance the real-time monitoring and precise control capabilities during substrate production, thereby improving the reliability of early monitoring and the stability of the production process.

[0090] Specifically, the intervention adjustment module is used to determine the amount of deviation in the image of the flexible substrate surface, and determining the reference time domain segment includes,

[0091] Used to determine the image similarity between the flexible substrate surface image and the corresponding standard image for each time domain segment;

[0092] The reciprocal of the image similarity is used to determine the deviation.

[0093] This is used to determine the time domain segment corresponding to the deviation as the reference time domain segment;

[0094] Among them, adjusting the substrate production parameters in the time domain dimension includes changing the substrate production parameters at predetermined intervals.

[0095] The preset time is set to 5 seconds to ensure that enough flexible substrate is produced for analysis. The production substrate parameters can be gradually changed after each preset time to quickly assess the production effect of different production substrate parameters.

[0096] In practice, the standard image is a pre-set reference image, such as a pattern sample that needs to be set on a flexible substrate. The pattern sample is used as the reference image. At the same time, there is no limitation on the method of determining the similarity of the images. It can be done by image processing algorithms or pattern recognition technology. It is only necessary to ensure that the similarity between images can be accurately evaluated. This will not be elaborated further.

[0097] Specifically, the intervention and adjustment module is used to determine the optimal substrate production parameters based on the reference time domain segment, including:

[0098] The substrate production parameters, which are gradient-adjusted within the reference time domain, are used to determine the optimal substrate production parameters.

[0099] This invention adjusts substrate production parameters in the time domain using a gradient, acquires flexible substrate surface images corresponding to each time segment, determines the deviation of the flexible substrate surface images, establishes a reference time segment, and determines the optimal substrate production parameters based on the reference time segment. In the closed-loop control process of high-speed flexible substrates, due to the complex dynamic characteristics of the production system, large-scale, one-time parameter adjustments can easily trigger oscillations. Therefore, a gradient adjustment strategy in the time domain is adopted. The core of this strategy is to transform the parameter adjustment action into an orderly, controlled, and observable system optimization process. Each adjustment step simultaneously acquires a substrate surface image produced under the influence of these parameters and calculates its deviation from the standard image to determine the reference time segment. The corresponding parameters are then determined as the optimal substrate production parameters. The aim is to transform experience-dependent, black-box parameter adjustment into an optimization based on online quality feedback, reducing production instability caused by parameter mutations, improving real-time monitoring and precise control capabilities in the substrate production process, and thus enhancing the reliability of early monitoring and the stability of the production process.

[0100] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A high-speed flexible substrate production monitoring system based on data analysis, characterized in that, include, The acquisition module includes a tension detection roller for acquiring tension in various regions of the flexible substrate, a laser ranging unit for detecting the spatial distance corresponding to the relative fixed height of each region, and an image acquisition unit for acquiring images of the surface of the flexible substrate. The pre-analysis module, which is connected to the acquisition module, is used to record the tension in each region in real time to determine the tension difference, determine the tension variation range based on the tension difference in each time domain segment, and determine the substrate variation region based on the difference between the relative distance thresholds of the corresponding spatial distances in each region. The verification module, which is connected to the pre-analysis module, is used to perform time variation verification based on the continuity of the tension variation range. In response to the result of the time variation verification, the spatial distance of each detected area within the tension variation range is extracted, a regional distance distribution map is constructed, and the deviation trend is determined based on the proportion of the substrate variation area in the regional distance distribution map. An intervention and adjustment module, which is connected to the acquisition module, the pre-analysis module, and the verification module respectively, adjusts the substrate production parameters in the time domain dimension in response to the deviation trend, acquires the surface images of the flexible substrate corresponding to each time domain segment, determines the deviation of the surface images of the flexible substrate, determines the reference time domain segment, and determines the optimal substrate production parameters based on the reference time domain segment. The substrate production parameters include the tension setting value and the conveying speed of the flexible substrate.

2. The high-speed flexible substrate production monitoring system based on data analysis according to claim 1, characterized in that, The pre-analysis module is used to record the tension in each region in real time to determine the tension difference. Used to determine the tension variance of each region detected at each time point within the time domain segment, and to determine the mean of the tension variance; This is used to determine the mean of the tension variance as the tension difference degree.

3. The high-speed flexible substrate production monitoring system based on data analysis according to claim 2, characterized in that, The pre-analysis module is used to determine the tension variation interval based on the tension difference within each time domain segment, including... If the tension difference within the time domain segment is greater than or equal to a predetermined tension difference threshold, then the time domain segment is determined to be a tension variation interval.

4. The high-speed flexible substrate production monitoring system based on data analysis according to claim 1, characterized in that, The pre-analysis module is used to determine the substrate variation regions based on the difference between the relative distance thresholds of the corresponding spatial distances in each region. Used to determine the difference between the relative distance thresholds corresponding to the spatial distances of each region; If the difference is greater than or equal to a preset difference threshold, the region is determined as a substrate variation region.

5. The high-speed flexible substrate production monitoring system based on data analysis according to claim 1, characterized in that, The verification module is used to perform time-varying verification based on the continuity of the tension variation range, including: This is used to determine whether the tension variation range meets the continuity condition. If the continuity condition is met, the time variation verification is passed. The continuity condition is that the tension variation intervals are continuous and the number of consecutive tension variation intervals is greater than a predetermined threshold.

6. The high-speed flexible substrate production monitoring system based on data analysis according to claim 1, characterized in that, The verification module is used to verify the results of time-varying responses, including... If the tension variation range passes the time variation verification, the spatial distance of each detected area within the tension variation range is extracted, and a regional distance distribution map is constructed. The deviation trend is determined based on the proportion of the substrate variation area in the regional distance distribution map.

7. The high-speed flexible substrate production monitoring system based on data analysis according to claim 6, characterized in that, The verification module is used to extract the spatial distance of each detected region within the tension variation range and construct a region distance distribution map, including... Used to determine the regional layout map corresponding to each region at a single moment; This is used to arrange the regional layout map according to the time sequence to obtain the regional distance distribution map; Used to mark substrate variation areas based on the regional distance distribution map.

8. The high-speed flexible substrate production monitoring system based on data analysis according to claim 1, characterized in that, The verification module is used to determine the deviation trend based on the proportion of substrate variation areas in the regional distance distribution map. Used to determine the proportion of substrate variation areas in the regional distance distribution map; If the percentage is greater than or equal to a predetermined percentage threshold, it is determined that the flexible substrate has a deviation trend.

9. The high-speed flexible substrate production monitoring system based on data analysis according to claim 1, characterized in that, The intervention adjustment module is used to determine the deviation of the flexible substrate surface image, and the determination of the reference time domain segment includes, Used to determine the image similarity between the flexible substrate surface image and the corresponding standard image for each time domain segment; The reciprocal of the image similarity is used to determine the deviation. This is used to determine the time domain segment corresponding to the deviation as the reference time domain segment; Among them, adjusting the substrate production parameters in the time domain dimension includes changing the substrate production parameters at predetermined intervals.

10. The high-speed flexible substrate production monitoring system based on data analysis according to claim 1, characterized in that, The intervention and adjustment module is used to determine the optimal substrate production parameters based on the reference time domain segment, including... The substrate production parameters, which are gradient-adjusted within the reference time domain, are used to determine the optimal substrate production parameters.

Images