A method for monitoring the thickness distribution uniformity of a coating

Abstract

This invention discloses a method for monitoring the uniformity of coating fabric thickness distribution, comprising: collecting the moisture content of each area of ​​the coating fabric through a humidity sensing module; collecting the angle between the fold ridge line and the scanning path and the density of folds covered in a single scan through a fold angle detection unit to form an initial data set; statistically analyzing the moisture content at the location of the wet area distribution; extracting the fold density and fold ridge line deflection direction of the secondary fold area from the updated monitoring data set; determining the tension matching direction by combining the angle and the moisture content at the location of the wet area distribution; verifying the effectiveness of threshold adjustment based on the density of folds and the fold density and fold ridge line deflection direction of the secondary fold area in the closed-loop monitoring cycle; and evaluating the thickness uniformity of the secondary fold area by combining the distribution pattern of random folds.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to a method for monitoring the uniformity of coating fabric thickness distribution. Background Technology

[0002] In industrial production, monitoring the uniformity of coated fabric thickness is a crucial technology, directly impacting product quality and production efficiency. As a fundamental material widely used in construction, packaging, and protective industries, the uniformity of coated fabric thickness not only affects the performance of the final product but also is closely related to cost control during the production process. Currently, although many monitoring methods are in use, they often reveal shortcomings when faced with complex production conditions.

[0003] A prior art patent, CN221781512U, discloses a coating thickness detection device for textile coating machines. It includes two rows of thickness detection probes disposed in a receiving groove. Multiple thickness detection probes in each row are arranged at intervals along a first direction. The thickness detection probes in the two rows are staggered along the first direction to reduce the detection blind zone. However, when the fabric surface is unstable or the humidity varies, the detection blind zone formed by a single row of probes makes it impossible to accurately determine the true thickness distribution.

[0004] Especially when the fabric surface is unstable or environmental factors are variable, monitoring results are easily interfered with when the fabric surface is affected by tension adjustment or humidity differences, making it impossible to accurately determine the true thickness distribution. This limitation makes it difficult to detect hidden problems in the production process in a timely manner. A deeper technical challenge lies in the interaction between fabric tension and humidity differences. Tension is an important means of smoothing the fabric surface and reducing wrinkles, but when the humidity of local areas of the fabric surface is uneven, the shrinkage degree of wet and dry areas is different, and tension may instead aggravate fabric deformation and form new wrinkles. These new wrinkles no longer exhibit regularity but are randomly distributed, seriously interfering with the signal acquisition and data analysis of monitoring equipment. Humidity differences, as a key factor affecting the fabric surface condition, have a complex and unpredictable mechanism, directly leading to a decrease in monitoring accuracy. Specifically, in actual production, after coating treatment, coated fabrics often have localized wet or dry areas due to uneven humidity distribution. For example, on a production line, the middle part of the fabric may contain more moisture because the coating is not yet dry, while the edge area is almost dry. When a large tension is applied to try to flatten the fabric, the wet area shrinks less and the dry area shrinks more, which eventually leads to local wrinkling of the fabric and affects the accuracy of thickness measurement.

[0005] Therefore, how to effectively identify and address the random distribution of local wrinkles on the fabric surface under the interactive influence of humidity differences and tension adjustment has become a key issue in improving the monitoring accuracy of coating fabric thickness uniformity. Summary of the Invention

[0006] To address the shortcomings of existing technologies, the present invention aims to provide a method for monitoring the uniformity of coating fabric thickness distribution, which solves the problem that, in existing technologies, coated fabrics often exhibit localized wet or dry areas due to uneven humidity distribution after coating treatment.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] A method for monitoring the uniformity of coating fabric thickness distribution, comprising:

[0009] The moisture content of each area of ​​the coated fabric, the angle between the fold ridge line and the scanning path, and the density of the fold arrangement covered by a single scan are collected to form an initial data set.

[0010] The moisture content of each region in the initial dataset is compared and analyzed region by region. The boundaries between wet and dry regions are identified based on the sudden changes in moisture content between adjacent regions, and the distribution of wet regions is marked. The shrinkage rate of wet and dry regions is calculated by combining the density of folds covered by a single scan, and potential secondary fold induction points are obtained.

[0011] The distribution pattern of folds is analyzed based on the angle between the fold ridge line and the scanning path. Potential secondary fold induction points are matched with the distribution location of wet areas to obtain overlapping areas. The degree of change in the fold distribution pattern within the overlapping areas is identified, and areas at risk of filter failure are identified to determine the threshold for anomaly identification.

[0012] Assess the degree of change in the filter failure risk area. If the change exceeds the preset threshold, reduce the anomaly identification threshold to extract the secondary folded area and the deflection direction of its fold ridge. Integrate the distribution location of the wet area with the density of the fold arrangement covered by a single scan to generate an updated monitoring data set.

[0013] The humidity of the distribution location of the wet area is statistically analyzed. The fold density and fold ridge deflection direction of the secondary fold area are extracted from the updated monitoring data set. Combined with the angle and the humidity of the distribution location of the wet area, the direction of the span tension is determined.

[0014] Adjust the tension level according to the tension adaptation direction, and judge the aggregation or dissipation state of random folds after tension adjustment by combining the distribution pattern of random folds. When random folds are aggregated, reduce the tension to balance the shrinkage rate of wet and dry areas. Combine the boundary between wet and dry areas to re-obtain the adjusted moisture content and fold arrangement density.

[0015] Furthermore, the initial data set is formed by collecting the moisture content of each area of ​​the coated fabric, the angle between the fold ridges and the scanning path, and the density of folds covered in a single scan, including:

[0016] The surface of the coated fabric is scanned in sections by a humidity sensing module. The detection areas are divided along the horizontal and vertical directions. The moisture content of each detection area is collected, and the moisture content values ​​are associated with the location coordinates of the detection area to obtain the regional moisture content distribution data.

[0017] The fold angle detection unit scans the surface of the coated fabric along a preset scanning path to obtain the angle between the fold ridge line and the scanning direction. The density of the fold ridge line within the coverage area of ​​a single scan is counted. The angle and density are then summarized into the regional moisture content distribution data to form an initial data set.

[0018] Furthermore, the process involves a zone-by-zone comparative analysis of the moisture content in each region of the initial dataset. Based on the abrupt changes in moisture content between adjacent regions, the boundaries between wet and dry zones are identified and the distribution locations of wet zones are marked. The shrinkage rates of wet and dry zones are calculated by combining the density of folds covered in a single scan, thus obtaining potential secondary fold induction points, including:

[0019] The moisture content values ​​of each detection area are extracted from the initial dataset. The moisture content of adjacent detection areas is compared zone by zone according to the location coordinates. The location where the moisture content difference exceeds the sudden change threshold is marked as the boundary between the wet and dry areas. The wet and dry areas are divided according to the boundary and the distribution location is marked.

[0020] The density of folds within the coverage area of ​​a single scan is obtained from the initial dataset. The number of fold ridges in the wet and dry areas is counted separately. The shrinkage rate is determined based on the number of fold ridges, and the shrinkage rates of the wet and dry areas are obtained.

[0021] The difference between the shrinkage rate of the wet area and the shrinkage rate of the dry area is calculated. Areas where the difference in shrinkage rate exceeds a threshold are selected at the boundary between the wet and dry areas, and these selected areas are identified as potential secondary wrinkle induction points.

[0022] Furthermore, the step of analyzing the distribution pattern of folds based on the angle between the fold ridge line and the scanning path, matching potential secondary fold induction points with the distribution location of wet areas to obtain overlapping regions, identifying the degree of change in fold distribution patterns within the overlapping regions, identifying areas at risk of filter failure, and determining anomaly identification thresholds includes:

[0023] The angle between the fold ridges and the scanning path in each detection area is extracted from the initial dataset. The folds are classified according to the numerical distribution characteristics of the angle. When the angles of the fold ridges in the same area are evenly distributed, they are marked as periodic fold distribution patterns. When the angles are discretely distributed, they are marked as random fold distribution patterns. The fold distribution pattern labels of each detection area are obtained.

[0024] The location coordinates of potential secondary fold induction points and the location coordinates of wet area distribution points are obtained, and spatial overlay comparison is performed to extract the areas where the location coordinates overlap as the overlapping areas.

[0025] For overlapping regions, read the fold distribution pattern markers and count the ratio of the number of random fold distribution pattern markers to the total number of markers as the degree of pattern transformation. When the degree of pattern transformation exceeds the preset transformation threshold, the overlapping region is marked as a filter failure risk region.

[0026] Anomaly identification thresholds are set based on the degree of mode shift within the filter failure risk area. The degree of mode shift and the anomaly identification threshold are inversely related, thus obtaining a matching anomaly identification threshold.

[0027] Furthermore, the assessment of the degree of change in the filter failure risk area, when exceeding a preset threshold, lowers the anomaly identification threshold to extract the secondary fold region and its fold ridge deflection direction, integrates the wet area distribution location with the density of fold arrangement covered by a single scan, and generates an updated monitoring data set, including:

[0028] Calculate the mode transition degree within the filter failure risk area, compare the mode transition degree with the preset risk threshold, and reduce the anomaly identification threshold when it exceeds the risk threshold to obtain the adjusted anomaly identification threshold.

[0029] The adjusted anomaly identification threshold is used to screen the ridge lines in the filter failure risk area. The position where the difference in the angle between adjacent ridge lines exceeds the deviation range is identified as the secondary ridge area. The change in the angle between each ridge line and the scanning path in the secondary ridge area is read to obtain the ridge line deflection direction.

[0030] The secondary folded areas, the fold ridge deflection direction, the location of the wet area, and the density of the fold arrangement covered by a single scan are correlated according to the location coordinates, and the correlated data is stored to form an updated monitoring data set.

[0031] Furthermore, the moisture content at the location of the statistically analyzed wet area is used to extract the fold density and fold ridge deflection direction of the secondary fold region from the updated monitoring data set. Combined with the angle of inclination and the moisture content at the location of the wet area, the direction of the stretching tension is determined, including:

[0032] Read the moisture content values ​​corresponding to the location of the wet area from the updated monitoring data set, perform moisture content statistics for each monitoring point within the location of the wet area, and obtain the average moisture content of the wet area.

[0033] The number of fold ridges in the secondary fold area is extracted from the updated monitoring data set. The fold density is obtained based on the ratio of the number of fold ridges to the number of detection points. The deflection direction and angle of each fold ridge in the secondary fold area are obtained. The proportion of clockwise and counterclockwise deflection is counted to obtain the dominant deflection direction.

[0034] The correlation between the dominant deflection direction, fold density, and average moisture content of the wet area is determined. When the dominant deflection direction is clockwise and the fold density exceeds the density threshold, the expansion tension adaptation direction is determined to be lateral increase. When the dominant deflection direction is counterclockwise and the average moisture content of the wet area exceeds the humidity threshold, the expansion tension adaptation direction is determined to be longitudinal decrease.

[0035] Furthermore, the step of adjusting the tension level according to the tension adaptation direction, judging the aggregation or dissipation state of random folds after tension adjustment based on the distribution pattern of random folds, reducing tension when random folds aggregate to balance the shrinkage rate of wet and dry areas, and re-obtaining the adjusted moisture content and fold density based on the boundary between wet and dry areas, includes:

[0036] The span tension level is adjusted according to the span tension adaptation direction. When the span tension adaptation direction is to increase laterally, the lateral span tension value is increased; when the span tension adaptation direction is to decrease longitudinally, the longitudinal span tension value is decreased, thus obtaining the adjusted span tension level.

[0037] Under the adjusted span tension level, the positional distribution of fold ridges within the random fold distribution pattern area is obtained. The distance between adjacent fold ridges is compared with the initial distance. A decrease in the distance is determined to be a clustered state, and an increase in the distance is determined to be a dissipated state.

[0038] For areas in a state of aggregation, the tension level is lowered to balance the shrinkage rates of the wet and dry areas, resulting in a balanced tension level.

[0039] Under the effect of the balanced span tension level, the moisture content and fold density are re-collected along the boundary between the wet and dry areas and updated to the monitoring data set, forming a closed-loop monitoring cycle.

[0040] Furthermore, the method also includes:

[0041] The effectiveness of threshold adjustment is verified by the density of fold arrangement, the fold density of secondary fold region, and the deflection direction of fold ridge line. The thickness uniformity of secondary fold region is evaluated by combining the distribution pattern of random folds.

[0042] Furthermore, the effectiveness of the threshold adjustment is verified based on the density of fold arrangement, the fold density in the secondary fold region, and the fold ridge deflection direction. Combined with the distribution pattern of random folds, the thickness uniformity of the secondary fold region is evaluated, including:

[0043] The updated fold density data is obtained from the closed-loop monitoring cycle. Fold density and fold ridge deflection direction are extracted from the secondary fold region. The fold density is compared with the fold density before the closed loop. When the fold density decreases and the dispersion of the fold ridge deflection direction is lower than the dispersion threshold, the threshold adjustment is determined to be effective, and the threshold effectiveness label is obtained.

[0044] The number of fold ridges in the random fold distribution pattern within the secondary fold region is read according to the threshold validity marker. The number of fold ridges in the random fold distribution pattern is compared with the total number of fold ridges to obtain the proportion of random folds.

[0045] The evaluation is based on the proportion of random folds and the dispersion of the fold ridge deflection direction. If the proportion of random folds is lower than the proportion threshold and the dispersion is lower than the dispersion threshold, the uniformity of the thickness of the secondary fold area is considered acceptable; otherwise, it is considered unacceptable.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] This invention discloses a method for monitoring the uniformity of coated fabric thickness distribution. It collects moisture content and wrinkle distribution data of the coated fabric through a humidity sensing module and a wrinkle angle detection unit. Addressing the business scenario problem of secondary wrinkles and uneven thickness caused by differences in shrinkage rates between wet and dry areas, it innovatively integrates a logically related solution of moisture content boundary identification, wrinkle distribution pattern analysis, and dynamic tension adjustment. First, this invention identifies the wet-dry boundary by comparing moisture content zone by zone, calculates the shrinkage rate based on wrinkle density, and predicts the induction point of secondary wrinkles. Then, it matches the wrinkle distribution with the location of the wet area, identifies the risk of filter failure, dynamically adjusts the threshold, and extracts secondary wrinkle features. Finally, it adjusts the tension level according to the tension adaptation direction to balance the shrinkage rate difference, and verifies the effectiveness of the threshold and the uniformity of thickness through closed-loop monitoring. Its core lies in the precise matching of dynamic tension adjustment and wrinkle distribution patterns, effectively reducing the risk of random wrinkle aggregation and significantly improving the uniformity of coated fabric thickness distribution, providing reliable technical support for industrial production. Attached Figure Description

[0048] Figure 1 This is a flowchart of a method for monitoring the uniformity of coating fabric thickness distribution in an embodiment of this solution;

[0049] Figure 2 This is a schematic diagram of a method for monitoring the uniformity of coating fabric thickness distribution in an embodiment of this solution;

[0050] Figure 3 This is another schematic diagram of a method for monitoring the uniformity of coating fabric thickness distribution in this embodiment. Detailed Implementation

[0051] The present invention will now be described in detail through specific embodiments:

[0052] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0053] like Figures 1-3 The method for monitoring the uniformity of coating fabric thickness distribution in this embodiment may specifically include:

[0054] S101 collects the moisture content of each area of ​​the coated fabric through the humidity sensing module, and collects the angle between the fold ridge line and the scanning path and the density of the fold arrangement covered by a single scan through the fold angle detection unit, thereby forming an initial data set.

[0055] In this step, the humidity sensing module performs a zoned scan of the coated fabric surface, dividing it into several detection areas along the horizontal and vertical directions, and collecting the moisture content within each detection area. The system then associates and records each moisture content value with the corresponding location coordinates of the detection area to obtain the regional moisture content distribution data.

[0056] Simultaneously, the wrinkle angle detection unit scans the surface of the coated fabric along a preset scanning path, obtaining the angle between the wrinkle ridges and the scanning direction along the scanning path, and calculating the density of the wrinkle ridges within the coverage area of ​​a single scan. The angle values ​​and density are then summarized into the regional moisture content distribution data to form an initial data set. During the monitoring of the coating fabric thickness uniformity, the humidity sensing module uses a high-sensitivity capacitive humidity sensor array to perform close-range non-contact scanning of the fabric surface.

[0057] Regarding the detection area division described in this application, in a preferred embodiment, a sensor collection point is set at fixed intervals along the transverse direction of the fabric surface, and the same spacing is applied longitudinally to form a grid-like detection area. The moisture content of each detection area is obtained by sensing changes in the dielectric constant of the fabric surface through the sensor; a higher moisture content value indicates a greater moisture content in the coating of that area.

[0058] In one possible implementation, the position coordinates of each detection area are recorded in two-dimensional coordinate form, with the horizontal coordinate representing the position in the fabric width direction and the vertical coordinate representing the position in the fabric travel direction. The moisture content values ​​are associated with and stored with the corresponding coordinates to form regional moisture content distribution data.

[0059] Specifically, the wrinkle angle detection unit uses a line laser scanning method to detect the fabric surface along a preset scanning path. The scanning path is set perpendicular to the direction of fabric travel. When the laser line illuminates the wrinkle ridge, a refraction offset occurs, and this offset amount d is captured by an image sensor, with the unit being pixels.

[0060] Further, the angle θ between the fold ridge line and the scanning direction is calculated. The specific calculation process is as follows: Where h is the fixed distance from the laser to the sensor in millimeters, θ is the angle between the fold ridge and the scanning direction, and arctan is the arctangent function. The angle reflects the orientation characteristics of the folds; the larger the angle, the greater the deviation between the fold ridge and the scanning path.

[0061] It should be noted that the density of the fold ridges within the coverage area of ​​a single scan is obtained by counting the number of fold ridges per unit length; a higher number of ridges indicates a denser fold arrangement. The angle, density, and moisture content distribution data of the area are summarized and stored according to their corresponding fabric positions to form an initial dataset containing both humidity and fold morphology information.

[0062] S102, perform a zone-by-zone comparative analysis of the moisture content of each region in the initial dataset, identify the boundary between wet and dry regions based on the sudden change in moisture content between adjacent regions and mark the distribution location of wet regions, and calculate the shrinkage rate of wet and dry regions under the action of span tension by combining the density of folds covered by a single scan, and obtain the potential secondary fold induction point under tension adjustment.

[0063] The moisture content values ​​of each detection area are extracted from the initial dataset, and the moisture content of adjacent detection areas is compared zone by zone according to the coordinates of the fabric position. If the difference in moisture content between two adjacent detection areas exceeds a preset threshold for sudden change, the position is marked as the boundary between the wet and dry areas. The wet and dry areas are then divided according to the boundary and their respective distribution positions are marked.

[0064] In the process of monitoring the uniformity of coated fabric thickness, identifying the boundary between wet and dry areas is a crucial step in determining the fabric's condition. For example, when the difference in moisture content between two adjacent detection areas exceeds 15%, it is determined that there is a significant abrupt change in humidity at that location, and this location is marked as the boundary between the wet and dry areas.

[0065] Specifically, the moisture content values ​​of each detection area are read sequentially according to the coordinates of the fabric surface. The difference between the moisture content of the current detection area and the moisture content of the adjacent detection areas is calculated. If the absolute value of the difference exceeds a preset threshold, a boundary marker is inserted at that location. Based on the boundary markers, the fabric surface is divided into several wet and dry zones. Areas with a moisture content higher than the average moisture content of the adjacent areas to the left and right of the boundary marker are classified as wet zones, and areas with a moisture content lower than the average moisture content are classified as dry zones.

[0066] The density of folds within the coverage area of ​​a single scan is obtained from the initial dataset. The number of fold ridges in the wet and dry areas is counted separately. The degree of shrinkage is determined based on the number of fold ridges, resulting in the wet area shrinkage rate and the dry area shrinkage rate. The difference between the wet area shrinkage rate and the dry area shrinkage rate is calculated. Regions with a shrinkage rate difference exceeding a preset threshold are selected at the boundary between the wet and dry areas. These selected regions are identified as potential secondary fold induction points under tension adjustment.

[0067] In one possible implementation, there is an inverse relationship between the number of pleat ridges and the degree of shrinkage. In areas with greater shrinkage, the fabric fibers tend to retract after the stretching tension is released, and the pleat ridges regroup, resulting in a larger number of ridges remaining. In areas with less shrinkage, the fabric fibers remain stretched under the stretching tension, and the pleat ridges are stretched and reduced, resulting in a smaller number of ridges remaining.

[0068] Based on this principle, the relative shrinkage rate can be determined by counting the number of pleats in the wet and dry areas. Areas with more pleats have a higher shrinkage rate, while areas with fewer pleats have a lower shrinkage rate. Typically, in the wet area, due to higher moisture content, the fibers are in an expanded state and easily stretch under the stretching tension, resulting in fewer pleats than in the dry area, indicating a lower shrinkage rate. In the dry area, due to lower moisture content, the fibers are in a contracted state and easily shrink after the stretching tension is released, resulting in more pleats than in the wet area, indicating a higher shrinkage rate.

[0069] Furthermore, at the boundary between the wet and dry zones, due to the difference in shrinkage on both sides, the direction of the stretching tension generates a shear effect at the boundary. When the difference between the wet and dry shrinkage rates exceeds a preset threshold, the shear effect at the boundary intensifies, and new wrinkles easily form on the surface of the coated fabric. The boundary area where the shrinkage difference exceeds the preset threshold is marked as a potential secondary wrinkle induction point. The location of this induction point reflects a high-risk area where the fabric surface is prone to secondary wrinkles during tension adjustment.

[0070] S103. Analyze the distribution pattern of random folds based on the angle between the fold ridges and the scanning path in the initial data set. Match the potential secondary fold induction points with the distribution location of the wet area to obtain the overlapping area. Identify the filter failure risk area and determine the anomaly identification threshold based on the degree of change of the fold distribution pattern from periodic to random within the overlapping area.

[0071] The angles between the ridge lines and the scanning path within each detection region are extracted from the initial dataset, and the folds are classified according to the numerical distribution characteristics of these angles. If the angles of the ridge lines within the same region are evenly spaced, the region is determined to be a periodic fold distribution pattern. If the angles are discretely distributed, the region is determined to be a random fold distribution pattern, thus obtaining a fold distribution pattern label for each detection region.

[0072] In the process of monitoring the uniformity of coated fabric thickness, determining the fold distribution pattern is a key step in identifying abnormal fabric conditions. For example, the angle between the fold ridge line and the scanning path reflects the direction characteristics of the folds. When the fabric surface is in a normal state, the folds usually exhibit a regular arrangement, and there is a fixed interval between the angles of each fold ridge line.

[0073] Specifically, the determination of a periodic fold distribution pattern is based on the fact that the angles between the fold ridges within the same detection area are evenly spaced. For example, if several fold ridges are collected within a certain detection area, and their angles with the scanning path are 15 degrees, 30 degrees, 45 degrees, and 60 degrees respectively, with the difference between any two adjacent angles being 15 degrees, then the folds in that area are determined to exhibit a periodic distribution pattern. The determination of a random fold distribution pattern is based on the fact that the angles are discretely distributed, and the difference between any two adjacent angles does not follow a fixed pattern, but varies randomly.

[0074] The location coordinates of potential secondary fold induction points and the location coordinates of wet area distribution points are obtained. The potential secondary fold induction points and the location coordinates of the wet area distribution points are then spatially superimposed and compared, and the areas where their location coordinates overlap are extracted as the coincident areas. In one possible implementation, the spatial superposition and comparison process involves projecting the location coordinates of the potential secondary fold induction points and the location coordinates of the wet area distribution points onto the same coordinate plane, and comparing the overlap of the two sets of coordinates point by point.

[0075] When the coordinates of a potential secondary fold induction point fall within the boundary of the wet zone distribution, that location is marked as a coincident point. The continuous area formed by all coincident points is the overlapping region. It should be noted that the existence of the overlapping region indicates that the location simultaneously meets two conditions: first, it is near the boundary where the shrinkage rate of the wet and dry zones differs significantly; second, it is located within the wet zone with a high moisture content.

[0076] The combination of these two conditions makes the overlapping area the most unstable location in the fabric surface, where the wrinkle pattern is more likely to change from periodic to random. For this overlapping area, the wrinkle distribution pattern markers within that area are read, and the ratio of the number of markers for random wrinkle distribution patterns to the total number of markers in that area is used as the degree of pattern transition. If the degree of pattern transition exceeds a preset transition threshold, the overlapping area is marked as a filter failure risk area.

[0077] Furthermore, the statistical method for determining the degree of pattern shift involves reading the fold distribution pattern markers of each detection point within the overlapping region, and then counting the number of detection points marked with a random fold distribution pattern relative to the total number of detection points in the region. The ratio of these two values ​​is used as a quantitative indicator of the degree of pattern shift. The numerical range of the degree of pattern shift is between zero and one; the closer the value is to one, the higher the proportion of random folds in the region.

[0078] In one embodiment, when the degree of mode transition exceeds a preset transition threshold, the overlapping area is determined to be a filter failure risk area. A filter failure risk area means that within this area, due to the highly random nature of the wrinkle morphology, conventional signal filtering methods struggle to accurately separate the wrinkle interference signal from the true thickness signal, leading to a decrease in the accuracy of thickness monitoring data. The preset transition threshold can be set according to the material characteristics of the coated fabric and the production process requirements, typically ranging from 0.6 to 0.8.

[0079] An anomaly identification threshold is set based on the degree of mode transition within the filter failure risk area. The degree of mode transition and the anomaly identification threshold are inversely related, resulting in an anomaly identification threshold that matches the filter failure risk area. It is understood that when the degree of mode transition within the filter failure risk area is high, it indicates strong randomness in the wrinkles of that area and a large amount of interference in the thickness monitoring signal. In this case, the anomaly identification threshold should be lowered to improve sensitivity to anomalous signals.

[0080] Preferably, the outlier detection threshold is set by linearly mapping the magnitude of the mode shift to a preset baseline threshold. The baseline threshold is an outlier detection standard determined experimentally when the fabric surface is in an ideal state, for example, set to 0.5. As the magnitude of the mode shift increases, the outlier detection threshold decreases proportionally. As the magnitude of the mode shift decreases, the outlier detection threshold increases proportionally. Through this setting method, the outlier detection threshold can be adaptively adjusted according to the actual state of the filter failure risk area.

[0081] S104, assess whether the degree of change in the filter failure risk area exceeds the preset threshold, and if it does, reduce the anomaly identification threshold to extract the secondary fold area and the deflection direction of its fold ridge, and then integrate the wet area distribution location with the density of the fold arrangement covered by a single scan to generate an updated monitoring data set.

[0082] During this process, the system calculates the Mode Transition Equation (MTE) within the filter failure risk area. The formula is: MTE equals the sum of the absolute values ​​of the changes in the angles of each ridge and fold, divided by the total number of ridges. This mode transition equation is compared with a preset risk threshold. If the mode transition equation exceeds the risk threshold, the anomaly identification threshold is reduced to a preset multiple of the original threshold, thus obtaining the adjusted anomaly identification threshold.

[0083] In the process of monitoring the uniformity of coated fabric thickness, the assessment of the filter failure risk area is the key basis for determining whether to initiate threshold adjustment. For example, the numerical value of the pattern transition degree is compared with a preset risk threshold. When the pattern transition degree exceeds the risk threshold, it indicates that the randomness of the wrinkle morphology in that area has reached a high level, and conventional anomaly identification thresholds are insufficient to accurately capture abnormal signals. Specifically, the anomaly identification threshold is reduced by multiplying the original threshold by a preset factor less than one. For example, if the original anomaly identification threshold is a fixed value and the preset factor is 0.7, then the adjusted anomaly identification threshold is 0.7 times the original threshold. The preset factor can be set according to the production process characteristics of the coated fabric; the smaller the factor, the greater the threshold reduction and the higher the sensitivity to abnormal signals.

[0084] The adjusted anomaly detection threshold is used to filter ridges within the filter failure risk area, specifically determining whether the difference in the angles between adjacent ridges exceeds a deviation range of 0.5% set by the adjusted anomaly detection threshold. Locations exceeding this deviation range are extracted as secondary ridge regions. For each ridge within this region, the change in its angle relative to the scanning path is read to obtain the ridge deflection direction. In one possible implementation, the determination criterion is the degree to which the angles of the ridges deviate from a periodic distribution pattern. A periodic distribution pattern refers to the equal spacing between the angles of adjacent ridges. When the difference between the angle of a ridge and its adjacent ridges exceeds a preset deviation range of this equal spacing, the ridge is determined to deviate from the periodic distribution pattern, and its location is marked as part of the secondary ridge region.

[0085] It should be noted that the deflection direction of the fold ridge lines is obtained by reading the change in the angle between each fold ridge line and the scanning path within the secondary fold region. The change in angle refers to the difference between the current angle of the fold ridge line and the angle recorded at that position in the previous scanning cycle. A positive difference indicates that the fold ridge line deflects clockwise, and a negative difference indicates that the fold ridge line deflects counterclockwise. Furthermore, the position coordinates of the secondary fold region, the deflection direction of the fold ridge lines, the distribution location of the wet area, and the density of the folds are correlated according to the transverse and longitudinal coordinates of the fabric surface, forming a data record structure indexed by position coordinates. The correlated data is stored according to the acquisition time, forming an updated monitoring data set. This set includes the spatial distribution and morphological change characteristics of secondary folds within the filter failure risk area.

[0086] S105, statistically analyze the moisture content at the location of the wet zone, extract the fold density and fold ridge deflection direction of the secondary fold area from the updated monitoring data set, and determine the expansion tension matching direction by combining the angle and the moisture content at the location of the wet zone.

[0087] The moisture content values ​​corresponding to the locations of the wet areas are read from the updated monitoring dataset. Moisture content statistics are then performed at each monitoring point within these locations to obtain the average moisture content of the wet area. This average reflects the overall moisture content level of the wet area. Simultaneously, the number of fold ridges within the secondary fold region is extracted from the updated monitoring dataset. The fold density is obtained based on the ratio of the number of fold ridges to the number of monitoring points in the secondary fold region.

[0088] Specifically, the wrinkle density is calculated by statistically analyzing the ratio of the number of wrinkle ridges within a secondary wrinkle region to the number of detection points within that region. For example, if twelve wrinkle ridges are detected within a secondary wrinkle region, and the region contains forty detection points, then the wrinkle density is 0.3. A higher wrinkle density indicates a denser distribution of wrinkles per unit area and a more unstable fabric surface. The deflection direction and corresponding angle of each wrinkle ridge within the secondary wrinkle region are obtained, and the proportion of clockwise and counterclockwise deflections is statistically analyzed to determine the dominant deflection direction. The determination of the dominant deflection direction is based on the statistical results of the deflection directions of each wrinkle ridge within the secondary wrinkle region; the direction with the larger proportion is the dominant deflection direction.

[0089] The determination is based on the correlation between the dominant deflection direction, pleat density, and average moisture content of the wet area. If the dominant deflection direction is clockwise and the pleat density exceeds a preset density threshold, the tension adaptation direction is determined to be lateral increase. If the dominant deflection direction is counterclockwise and the average moisture content of the wet area exceeds a preset humidity threshold, the tension adaptation direction is determined to be longitudinal decrease. It should be noted that the determination of the tension adaptation direction is the result of a comprehensive correlation between the dominant deflection direction, pleat density, and average moisture content of the wet area. When the dominant deflection direction is clockwise, it indicates that the pleat ridge line is deflected laterally towards the fabric surface. In this case, if the pleat density is high, the lateral tension should be increased to smooth the lateral pleats. When the dominant deflection direction is counterclockwise, it indicates that the pleat ridge line is deflected longitudinally towards the fabric surface. In this case, if the moisture content of the wet area is too high, causing uneven longitudinal shrinkage, the tension should be decreased longitudinally to balance the longitudinal shrinkage.

[0090] S106, adjust the tension level according to the tension adaptation direction, and judge the aggregation or dissipation state of random folds after tension adjustment by combining the distribution pattern of random folds. If random folds are aggregated, reduce the tension to balance the shrinkage rate of wet and dry areas. Combine the boundary between wet and dry areas to re-obtain the adjusted moisture content and fold arrangement density to form a closed-loop monitoring cycle.

[0091] The fabric tension level is adjusted according to the direction of the tension adaptation. If the adaptation direction is to increase laterally, the fabric tension value in the lateral direction is increased. If the adaptation direction is to decrease longitudinally, the fabric tension value in the longitudinal direction is decreased, thus obtaining the adjusted fabric tension level.

[0092] Under the adjusted tension level, the positional distribution of each crease ridge within the random crease distribution pattern area is obtained. The initially calculated spacing between adjacent crease ridges is taken from the monitoring data set as the spacing before adjustment, and the current spacing is compared with it. If the spacing decreases, the random creases are determined to be in a clustered state; if the spacing increases, the random creases are determined to be in a dissipated state. Specifically, the determination of the clustered state of random creases is based on a decreasing trend in the spacing between adjacent crease ridges. The clustered state indicates that the tension adjustment has failed to smooth out the creases, instead causing the creases to concentrate in a certain area, resulting in a decrease in the smoothness of the fabric surface. Conversely, the dissipated state indicates that the tension adjustment has gradually smoothed out the creases, improving the smoothness of the fabric surface.

[0093] For areas where random wrinkles are clustered, the tension level is reduced by a preset margin to balance the shrinkage rates of the wet and dry areas, resulting in a balanced tension level. The principle behind reducing tension is to decrease the stretching effect on the fabric, making the shrinkage rates of the wet and dry areas more consistent. Specifically, the shrinkage rate is quantified by the relationship between the density of wrinkles and the area. The shrinkage rate R is equal to the number of wrinkle ridges in the area divided by the product of the area's length along the scanning path and its width in the vertical direction. A higher shrinkage rate indicates more wrinkles remaining per unit area, and a stronger tendency for shrinkage after the tension is released. The shrinkage rates of the wet and dry areas are calculated separately to obtain quantified values ​​for the shrinkage rates of both areas.

[0094] In one embodiment, the balanced tension level is a value adjusted to achieve a similar shrinkage rate between the wet and dry areas. This tension level maintains the basic tension of the fabric while preventing excessive shrinkage differences due to excessive tension. Under the balanced tension level, the moisture content and fold density of each detection area are re-collected along the boundary between the wet and dry areas. The re-collected data is then updated into the monitoring dataset, forming a closed-loop monitoring cycle. This closed-loop mechanism allows the monitoring process to dynamically adjust according to real-time changes in the fabric's condition, continuously optimizing the tension level to maintain the fabric's smoothness.

[0095] S107. Based on the density of fold arrangement and the fold density and fold ridge deflection direction in the secondary fold region during closed-loop monitoring, the effectiveness of threshold adjustment is verified, and the thickness uniformity of the secondary fold region is evaluated in combination with the distribution pattern of random folds.

[0096] The updated fold density data is obtained from the closed-loop monitoring cycle. Fold density and fold ridge deflection orientation are extracted from the secondary fold region. The fold density is then compared with that before the closed-loop monitoring cycle. If the fold density decreases and the dispersion of the fold ridge deflection orientation is lower than a preset dispersion threshold, the threshold adjustment is deemed effective, and a threshold effectiveness marker is obtained.

[0097] Specifically, the dispersion of the deflection azimuth of fold ridges refers to the distribution of the deflection azimuth values ​​among the individual fold ridge deflection azimuth values ​​within the secondary fold region. The dispersion is calculated by statistically analyzing the absolute value of the deviation between each fold ridge deflection azimuth value and the average deflection azimuth value, summing all absolute deviations, and then dividing by the total number of fold ridges to obtain the dispersion value. If the fold density shows a decreasing trend compared to before the closed-loop monitoring cycle, and the dispersion is lower than the preset dispersion threshold, then the anomaly identification threshold adjustment performed in the previous step is deemed effective, and the threshold validity is recorded and marked as valid.

[0098] The number of fold ridges in the random fold distribution pattern within the secondary fold region is read based on the threshold validity marker. This number is then compared to the total number of fold ridges within the secondary fold region to obtain the random fold proportion. For example, if twenty fold ridges are detected within a secondary fold region, and six of them belong to the random fold distribution pattern, the random fold proportion is 0.3%. Furthermore, the thickness uniformity assessment is based on a comprehensive evaluation of the random fold proportion and the degree of dispersion.

[0099] When the proportion of random folds is lower than a preset proportion threshold and the dispersion is lower than a preset dispersion threshold, it indicates that the fold distribution within the secondary fold region is relatively regular, and the folds have little interference with thickness measurement; therefore, the thickness uniformity of this region is considered acceptable. When the proportion of random folds is higher than the proportion threshold or the dispersion is higher than the dispersion threshold, it indicates that the fold distribution is disordered, and the thickness uniformity is considered unacceptable. Regarding the collection and processing of personal information in this application, explicit consent from individuals has been obtained in accordance with relevant laws and regulations, ensuring that all personal information processing activities are based on legal authorization and fully comply with national compliance requirements for personal information protection.

[0100] If the technical solution of this application involves the processing of personal information, the relevant products have established a sound user authorization mechanism: before collecting, using, or sharing personal information, the obligation to inform is fulfilled in accordance with the law, and the individual's voluntary and explicit consent is obtained; if sensitive personal information is involved, the user's separate and explicit consent is further obtained. Specific measures include, but are not limited to: setting up prominent prompts in the information collection area, or clearly displaying the processing rules (including the processor, purpose, method, information type, etc.) through electronic interfaces such as pop-ups, checkboxes, and active submissions, to ensure that users voluntarily authorize based on their knowledge. All personal information processing activities strictly comply with national laws and regulations, especially the relevant provisions of the "Personal Information Protection Law of the People's Republic of China," to effectively safeguard the legitimate rights and interests of personal information subjects.

[0101] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method of monitoring the thickness profile uniformity of a coated fabric, characterized by, include: The moisture content of each area of ​​the coated fabric, the angle between the fold ridge line and the scanning path, and the density of the fold arrangement covered by a single scan are collected to form an initial data set. The moisture content of each region in the initial dataset is compared and analyzed region by region. The boundaries between wet and dry regions are identified based on the sudden changes in moisture content between adjacent regions, and the distribution of wet regions is marked. The shrinkage rate of wet and dry regions is calculated by combining the density of folds covered by a single scan, and potential secondary fold induction points are obtained. The distribution pattern of folds is analyzed based on the angle between the fold ridge line and the scanning path. Potential secondary fold induction points are matched with the distribution location of wet areas to obtain overlapping areas. The degree of change in the fold distribution pattern within the overlapping areas is identified, and areas at risk of filter failure are identified to determine the threshold for anomaly identification. Assess the degree of change in the filter failure risk area. If the change exceeds the preset threshold, reduce the anomaly identification threshold to extract the secondary folded area and the deflection direction of its fold ridge. Integrate the distribution location of the wet area with the density of the fold arrangement covered by a single scan to generate an updated monitoring data set. The humidity of the distribution location of the wet area is statistically analyzed. The fold density and fold ridge deflection direction of the secondary fold area are extracted from the updated monitoring data set. Combined with the angle and the humidity of the distribution location of the wet area, the direction of the span tension is determined. Adjust the tension level according to the tension adaptation direction, and judge the aggregation or dissipation state of random folds after tension adjustment by combining the distribution pattern of random folds. When random folds are aggregated, reduce the tension to balance the shrinkage rate of wet and dry areas. Combine the boundary between wet and dry areas to re-obtain the adjusted moisture content and fold arrangement density.

2. The coating thickness profile uniformity monitoring method according to claim 1, wherein, The initial data set is formed by collecting the moisture content of each area of ​​the coated fabric, the angle between the fold ridges and the scanning path, and the density of folds covered in a single scan, including: The surface of the coated fabric is scanned in sections by a humidity sensing module. The detection areas are divided along the horizontal and vertical directions. The moisture content of each detection area is collected, and the moisture content values ​​are associated with the location coordinates of the detection area to obtain the regional moisture content distribution data. The fold angle detection unit scans the surface of the coated fabric along a preset scanning path to obtain the angle between the fold ridge line and the scanning direction. The density of the fold ridge line within the coverage area of ​​a single scan is counted. The angle and density are then summarized into the regional moisture content distribution data to form an initial data set.

3. The coating thickness profile uniformity monitoring method according to claim 1, wherein, The process involves comparing the moisture content of each region in the initial dataset region by region, identifying the boundaries between wet and dry regions based on abrupt changes in moisture content between adjacent regions, marking the distribution location of wet regions, and calculating the shrinkage rate of wet and dry regions based on the density of folds covered by a single scan. This yields potential secondary fold induction points, including: The moisture content values ​​of each detection area are extracted from the initial dataset. The moisture content of adjacent detection areas is compared zone by zone according to the location coordinates. The location where the moisture content difference exceeds the sudden change threshold is marked as the boundary between the wet and dry areas. The wet and dry areas are divided according to the boundary and the distribution location is marked. The density of folds within the coverage area of ​​a single scan is obtained from the initial dataset. The number of fold ridges in the wet and dry areas is counted separately. The shrinkage rate is determined based on the number of fold ridges, and the shrinkage rates of the wet and dry areas are obtained. The difference between the shrinkage rate of the wet area and the shrinkage rate of the dry area is calculated. Areas where the difference in shrinkage rate exceeds a threshold are selected at the boundary between the wet and dry areas, and these selected areas are identified as potential secondary wrinkle induction points.

4. The coating thickness profile uniformity monitoring method according to claim 1, wherein, The process involves analyzing the fold distribution pattern based on the angle between the fold ridge line and the scanning path, matching potential secondary fold induction points with the distribution location of wet areas to obtain overlapping regions, identifying the degree of change in fold distribution patterns within the overlapping regions, identifying areas at risk of filter failure, and determining anomaly identification thresholds, including: The angle between the fold ridges and the scanning path in each detection area is extracted from the initial dataset. The folds are classified according to the numerical distribution characteristics of the angle. When the angles of the fold ridges in the same area are evenly distributed, they are marked as periodic fold distribution patterns. When the angles are discretely distributed, they are marked as random fold distribution patterns. The fold distribution pattern labels of each detection area are obtained. The location coordinates of potential secondary fold induction points and the location coordinates of wet area distribution points are obtained, and spatial overlay comparison is performed to extract the areas where the location coordinates overlap as the overlapping areas. For overlapping regions, read the fold distribution pattern markers and count the ratio of the number of random fold distribution pattern markers to the total number of markers as the degree of pattern transformation. When the degree of pattern transformation exceeds the preset transformation threshold, the overlapping region is marked as a filter failure risk region. Anomaly identification thresholds are set based on the degree of mode shift within the filter failure risk area. The degree of mode shift and the anomaly identification threshold are inversely related, thus obtaining a matching anomaly identification threshold.

5. The method for monitoring the uniformity of coating fabric thickness distribution according to claim 1, characterized in that, The assessment evaluates the degree of change in the filter failure risk area. When the change exceeds a preset threshold, the anomaly identification threshold is lowered to extract the secondary fold region and its fold ridge deflection direction. The wet area distribution location and the density of fold arrangement covered by a single scan are integrated to generate an updated monitoring data set, including: Calculate the mode transition degree within the filter failure risk area, compare the mode transition degree with the preset risk threshold, and reduce the anomaly identification threshold when it exceeds the risk threshold to obtain the adjusted anomaly identification threshold. The adjusted anomaly identification threshold is used to screen the ridge lines in the filter failure risk area. The position where the difference in the angle between adjacent ridge lines exceeds the deviation range is identified as the secondary ridge area. The change in the angle between each ridge line and the scanning path in the secondary ridge area is read to obtain the ridge line deflection direction. The secondary folded areas, the fold ridge deflection direction, the location of the wet area, and the density of the fold arrangement covered by a single scan are correlated according to the location coordinates, and the correlated data is stored to form an updated monitoring data set.

6. The method for monitoring the uniformity of coating fabric thickness distribution according to claim 1, characterized in that, The moisture content at the location of the statistical wet area distribution is used to extract the fold density and fold ridge deflection direction of the secondary fold region from the updated monitoring data set. Combined with the angle of inclination and the moisture content at the location of the wet area distribution, the span tension adaptation direction is determined, including: Read the moisture content values ​​corresponding to the location of the wet area from the updated monitoring data set, perform moisture content statistics for each monitoring point within the location of the wet area, and obtain the average moisture content of the wet area. The number of fold ridges in the secondary fold area is extracted from the updated monitoring data set. The fold density is obtained based on the ratio of the number of fold ridges to the number of detection points. The deflection direction and angle of each fold ridge in the secondary fold area are obtained. The proportion of clockwise and counterclockwise deflection is counted to obtain the dominant deflection direction. The correlation between the dominant deflection direction, fold density, and average moisture content of the wet area is determined. When the dominant deflection direction is clockwise and the fold density exceeds the density threshold, the expansion tension adaptation direction is determined to be lateral increase. When the dominant deflection direction is counterclockwise and the average moisture content of the wet area exceeds the humidity threshold, the expansion tension adaptation direction is determined to be longitudinal decrease.

7. The method for monitoring the uniformity of coating fabric thickness distribution according to claim 1, characterized in that, The process of adjusting the expansion tension level according to the expansion tension adaptation direction, judging the aggregation or dissipation state of random folds after tension adjustment based on the distribution pattern of random folds, reducing tension when random folds aggregate to balance the shrinkage rate of wet and dry areas, and re-obtaining the adjusted moisture content and fold arrangement density based on the boundary between wet and dry areas includes: The span tension level is adjusted according to the span tension adaptation direction. When the span tension adaptation direction is to increase laterally, the lateral span tension value is increased; when the span tension adaptation direction is to decrease longitudinally, the longitudinal span tension value is decreased, thus obtaining the adjusted span tension level. Under the adjusted span tension level, the positional distribution of fold ridges within the random fold distribution pattern area is obtained. The distance between adjacent fold ridges is compared with the initial distance. A decrease in the distance is determined to be a clustered state, and an increase in the distance is determined to be a dissipated state. For areas in a state of aggregation, the tension level is lowered to balance the shrinkage rates of the wet and dry areas, resulting in a balanced tension level. Under the effect of the balanced span tension level, the moisture content and fold density are re-collected along the boundary between the wet and dry areas and updated to the monitoring data set, forming a closed-loop monitoring cycle.

8. The method for monitoring the uniformity of coating fabric thickness distribution according to claim 1, characterized in that, The method further includes: The effectiveness of threshold adjustment is verified by the density of fold arrangement, the fold density of secondary fold region, and the deflection direction of fold ridge line. The thickness uniformity of secondary fold region is evaluated by combining the distribution pattern of random folds.

9. The method for monitoring the uniformity of coating fabric thickness distribution according to claim 8, characterized in that, The effectiveness of the threshold adjustment is verified based on the density of fold arrangement, the fold density of the secondary fold region, and the deflection direction of the fold ridge line. Combined with the distribution pattern of random folds, the thickness uniformity of the secondary fold region is evaluated, including: The updated fold density data is obtained from the closed-loop monitoring cycle. Fold density and fold ridge deflection direction are extracted from the secondary fold region. The fold density is compared with the fold density before the closed loop. When the fold density decreases and the dispersion of the fold ridge deflection direction is lower than the dispersion threshold, the threshold adjustment is determined to be effective, and the threshold effectiveness label is obtained. The number of fold ridges in the random fold distribution pattern within the secondary fold region is read according to the threshold validity marker. The number of fold ridges in the random fold distribution pattern is compared with the total number of fold ridges to obtain the proportion of random folds. The evaluation is based on the proportion of random folds and the dispersion of the fold ridge deflection direction. If the proportion of random folds is lower than the proportion threshold and the dispersion is lower than the dispersion threshold, the uniformity of the thickness of the secondary fold area is considered acceptable; otherwise, it is considered unacceptable.

Images