Special paper slitting and embossing process thickness detection system

Abstract

The invention relates to the technical field of intelligent control, in particular to a specialty paper slitting and embossing process thickness detection system, which comprises an acquisition module used for determining a plurality of dynamic reference points at a preset position at the downstream of an outlet of an embossing roller, acquiring original displacement data in real time, constructing a dynamic polygon according to the dynamic reference points, and determining the thickness of the dynamic polygon; calculating a centroid offset and an area change rate, generating a dynamic adjustment coefficient, and calibrating the displacement original data through the dynamic adjustment coefficient to obtain calibrated displacement data; the filtering module is used for performing high-frequency filtering processing on the calibration displacement data to generate filtering displacement data; and the anti-interference module is used for performing anti-electromagnetic interference processing on the filtering displacement data to generate anti-interference displacement data. According to the invention, through accurate detection, anti-interference processing and real-time control of the whole link, high-precision monitoring and regulation of the thickness of the special paper in the slitting and embossing process are realized.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control technology, in particular to a thickness detection system for a special paper slitting and embossing process. Background Art

[0002] Specialty paper is widely used in packaging, electronics, medical and other fields, and its product quality directly affects the application effect of downstream industries; slitting and embossing are key production processes, and the uniformity of paper thickness directly determines product quality, which requires real-time and accurate detection and control.

[0003] However, current thickness detection mainly relies on contact sensors (such as mechanical thickness gauges) or non-contact sensors (such as laser displacement sensors), which may face multiple technical challenges in practical applications: Fixed reference points may be difficult to match the dynamic changes of paper during high-speed movement, and static errors may be easily introduced due to the relative position offset between the reference point and the paper; the high-frequency vibrations that may be generated by high-speed operation of the equipment will introduce oscillation noise, and the electromagnetic environment formed by equipment such as the embossing roller motor and the electronic control system will interfere with the sensor signal, resulting in pulse jumps or energy anomalies in the displacement data; the drift of paper and sensors caused by temperature and humidity fluctuations may be difficult to eliminate; when thickness deviations occur, some traditional systems rely on manual judgment or simple threshold trigger adjustments, which may lack refined analysis of deviation characteristics (such as direction, amplitude, distribution), have a single adjustment strategy, and the actuator response and thickness monitoring may lack real-time linkage, which may easily lead to excessive or insufficient adjustments, making it difficult to ensure thickness stability. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to overcome the shortcomings of the existing technology and provide a thickness detection system for the special paper slitting and embossing process. Through precise detection, anti-interference processing and real-time control of the entire link, high-precision monitoring and control of the thickness of the special paper slitting and embossing process are achieved.

[0005] In order to solve the above technical problems, the basic concept of the technical solution adopted by the present invention is: A thickness detection system for special paper slitting and embossing process, comprising: An acquisition module is used to determine multiple dynamic reference points at a preset position downstream of the embossing roller outlet, collect raw displacement data in real time, construct dynamic polygons based on the dynamic reference points, calculate the centroid offset and area change rate, and generate dynamic adjustment coefficients. The dynamic adjustment coefficients are used to calibrate the raw displacement data to obtain calibrated displacement data; A filtering module, configured to perform high-frequency filtering on the calibration displacement data to generate filtered displacement data; an anti-interference module, configured to apply anti-electromagnetic interference processing to the filtered displacement data to generate anti-interference displacement data; a correction module, configured to perform edge contour reconstruction based on a convex hull algorithm on the anti-interference displacement data and output edge reconstructed displacement data; a correction module, configured to perform temperature and humidity drift correction on the edge reconstruction displacement data to generate corrected displacement data; The alarm module is used to calculate the thickness value in real time based on the corrected displacement data, and to trigger an alarm signal immediately and synchronously adjust the embossing roller pressure or slitting tension when the thickness deviation exceeds a preset threshold.

[0006] In a preferred embodiment of any of the above schemes, a plurality of dynamic reference points are determined at a preset position downstream of the embossing roller outlet, raw displacement data is collected in real time, a dynamic polygon is constructed based on the dynamic reference points, the centroid offset and area change rate are calculated and a dynamic adjustment coefficient is generated, and the raw displacement data is calibrated using the dynamic adjustment coefficient to obtain calibrated displacement data, including: Dynamically determine multiple dynamic reference points at a preset distance downstream of the embossing roller outlet, and collect the original displacement data of each dynamic reference point in real time; Based on the displacement original data, a dynamic polygon is constructed with the dynamic reference point as a vertex, and the spatial coordinate sequence of each vertex is recorded in real time; Calculating the centroid coordinates of the dynamic polygon based on the spatial coordinate sequence, and comparing the centroid coordinates of the polygon with a preset reference centroid position to generate a centroid offset; and simultaneously calculating the area of ​​the current dynamic polygon, and comparing the area of ​​the dynamic polygon with a preset reference area to generate an area change rate; Converting the centroid offset into a distance scalar value, establishing a weighted function relationship between the distance scalar value and the area change rate, and outputting a dynamic adjustment coefficient; Based on the dynamic adjustment coefficient, the original displacement data is decomposed into independent components of each reference point, and a compensation value matching the dynamic adjustment coefficient is applied to each independent component; The compensated displacement data sequence is reconstructed to generate the calibration displacement data.

[0007] In a preferred embodiment of any of the above solutions, performing high-frequency filtering on the calibration displacement data to generate filtered displacement data includes: Based on the calibrated displacement data, high-frequency oscillation components in the displacement sequence are extracted, and excessive oscillation waves are filtered out according to a preset vibration amplitude threshold to generate time-domain noise-reduced displacement data; Performing frequency domain conversion processing on the time domain noise reduction displacement data to generate corresponding frequency domain energy distribution; Based on the frequency domain energy distribution and in combination with the production line running speed, a target frequency band range is determined, noise energy components outside the target frequency band range are attenuated, and frequency domain filtering data is generated; The frequency domain filtered data is subjected to time domain inverse conversion processing to be reconstructed into a time domain displacement sequence to generate the filtered displacement data.

[0008] In a preferred embodiment of any of the above solutions, applying anti-electromagnetic interference processing to the filtered displacement data to generate anti-interference displacement data includes: Based on the filtered displacement data, the current change rate of the embossing machine power supply is monitored in real time, and when the current change rate exceeds a preset mutation threshold, an interference event marker is activated and the corresponding time position is recorded; Based on the displacement data period corresponding to the interference event marker positioning, converting the displacement data of the period into a frequency domain energy spectrum; Identifying an abnormal energy peak in a preset interference frequency band in the frequency domain energy spectrum, applying a band-stop filtering process centered on the abnormal peak to suppress the abnormal peak energy within the spectrum baseline range, and generating band-stop filtered frequency domain data; Detecting a short-duration pulse jump signal in the band-stop filtered frequency domain data, performing window average smoothing processing on the detected pulse jump, and generating corrected frequency domain data; The corrected frequency domain data is inversely converted into a time domain signal sequence, reconstructed into continuous displacement data, and the anti-interference displacement data is generated.

[0009] In a preferred embodiment of any of the above solutions, performing edge contour reconstruction based on a convex hull algorithm on the anti-interference displacement data and outputting edge reconstructed displacement data includes: A two-dimensional structured grid coordinate system is established based on the anti-interference displacement data, wherein the coordinate system takes the center of the sensor measurement spot as the origin, and the grid unit size is determined according to the preset spot size parameters and the burr influence range; In the two-dimensional structured grid coordinate system, extracting spatial distribution data of burr fibers within a unit, constructing a convex hull boundary of the burr region based on the spatial distribution data, and calculating a light scattering attenuation weight factor according to the distance between the convex hull boundary and the unit center; Process the attenuation weight factors of all grid cells to generate a grid cell attenuation coefficient matrix; Mapping the anti-interference displacement data to a grid coordinate system, determining the grid unit to which each displacement data point belongs, calculating the displacement compensation value based on the attenuation coefficient of the unit, and correcting the original displacement data using the compensation value to obtain the compensated displacement data; Verification is performed on the edge area of ​​the compensated displacement data, a continuous displacement sequence is reconstructed based on the verified displacement data, and the edge reconstructed displacement data is output.

[0010] In a preferred embodiment of any of the above solutions, performing temperature and humidity drift correction on the edge reconstructed displacement data to generate corrected displacement data includes: Real-time collection of ambient temperature sensor data and ambient humidity sensor data; Calculate the difference between the current temperature and the reference temperature based on the temperature sensor data, and generate the original linear compensation component according to the preset temperature compensation coefficient; Calculate the difference between the current humidity and the reference humidity based on the humidity sensor data, and generate the original nonlinear compensation component according to the preset humidity compensation coefficient; Normalizing the original linear compensation component and the original nonlinear compensation component to obtain normalized linear compensation component and nonlinear compensation component; Based on the normalized linear compensation component and nonlinear compensation component, a basic composite value is generated through weighted fusion calculation; Generating a final composite compensation value according to the basic composite value and the time delay parameter compensation; Dynamically superimposing the final composite compensation value onto the edge reconstructed displacement data in a time series to generate preliminary corrected displacement data; The stability of the displacement data is verified under constant temperature and humidity conditions, and when the drift amount is lower than a preset drift tolerance threshold, the corrected displacement data is generated.

[0011] In a preferred embodiment of any of the above solutions, a thickness value is calculated in real time based on the corrected displacement data, and an alarm signal is immediately triggered when the thickness deviation exceeds a preset threshold value, and the embossing roller pressure or slitting tension is synchronously adjusted, including: Separating the upper and lower surface displacement components based on the corrected displacement data, and calculating the displacement differences at corresponding time points to generate a real-time thickness sequence; Compare the real-time thickness sequence with the preset standard thickness value and dynamically calculate the thickness deviation value; When the thickness deviation value exceeds the preset deviation tolerance, an over-limit alarm instruction and deviation characteristic parameters are generated; Executing a control decision based on the deviation characteristic parameter specifically includes: The direction of the actuator is determined according to the positive or negative sign of the deviation value, and a single-mechanism or dual-mechanism coordinated adjustment strategy is selected according to the deviation amplitude to generate an instruction group for the embossing roller pressure adjustment amount and the slitting tension adjustment amount; Instantly trigger a multi-modal alarm signal and synchronously send the command group to the actuator through a preset industrial communication protocol; The actuator feedback signal is obtained in real time, and when it is detected that the adjusted secondary thickness value returns to the standard range, the closed-loop correction is completed.

[0012] In a second aspect, a method for detecting thickness of specialty paper during slitting and embossing comprises the following steps: Determine multiple dynamic reference points at a preset position downstream of the embossing roller outlet, collect original displacement data in real time, construct dynamic polygons based on the dynamic reference points, calculate the centroid offset and area change rate, and generate dynamic adjustment coefficients. Use the dynamic adjustment coefficients to calibrate the original displacement data to obtain calibrated displacement data. performing high-frequency filtering on the calibration displacement data to generate filtered displacement data; applying anti-electromagnetic interference processing to the filtered displacement data to generate anti-interference displacement data; Performing edge contour reconstruction based on a convex hull algorithm on the anti-interference displacement data, and outputting edge reconstructed displacement data; performing temperature and humidity drift correction on the edge reconstructed displacement data to generate corrected displacement data; The thickness value is calculated in real time based on the corrected displacement data, and when the thickness deviation exceeds a preset threshold, an alarm signal is immediately triggered and the embossing roller pressure or slitting tension is adjusted synchronously.

[0013] After adopting the above technical solution, the present invention has the following beneficial effects compared with the prior art: The acquisition module uses dynamic fiducials and dynamic polygon analysis to avoid static errors from fixed fiducials, ensuring that the raw data accurately reflects the dynamic characteristics of the paper. The filtering module specifically filters out high-frequency oscillation noise, while the anti-interference module eliminates electromagnetic interference through electromagnetic interference event marking and band-stop filtering. The correction module accurately compensates for light scattering errors caused by burrs using a convex hull algorithm, and the correction module eliminates the effects of temperature and humidity drift. Throughout the data processing process, layer-by-layer noise reduction and correction are implemented to minimize the impact of various interferences on detection, ensuring that the final thickness data truly reflects the actual state of the paper. Dynamic fiducials adapt to changes in paper motion, while band-stop filtering and pulse smoothing mitigate electromagnetic interference. The convex hull algorithm corrects for burrs, and temperature and humidity drift correction offsets environmental fluctuations. This ensures that the detection system remains stable in production environments characterized by high-speed motion, dense electromagnetic fields, widespread burrs, and fluctuating temperature and humidity. When thickness exceeds the limit, it quickly triggers a multi-modal alarm and generates precise adjustment commands. Combined with actuator feedback and secondary thickness monitoring, this ensures that adjustments meet target requirements, preventing the accumulation of thickness anomalies that could lead to batch defects and improving product thickness consistency.

[0014] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation of the present application. Some specific embodiments of the present application will be described in detail in an illustrative and non-restrictive manner with reference to the drawings. The same reference numerals in the drawings indicate the same or similar components or parts. It should be understood by those skilled in the art that these drawings are not necessarily drawn to scale. In the drawings: Figure 1 It is a schematic diagram of the thickness detection system of the special paper slitting and embossing process of the present invention.

[0016] Figure 2 The present invention is a flow chart of a control method for a thickness detection system for a specialty paper slitting and embossing process. DETAILED DESCRIPTION

[0017] In order to enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this application.

[0018] The following embodiment of the present application takes the thickness detection system of the special paper slitting and embossing process as an example to explain the solution of the present application in detail, but this embodiment does not limit the scope of protection of the present application.

[0019] like Figure 1 As shown, the present invention provides a thickness detection system for a special paper slitting and embossing process, comprising: An acquisition module 11 is configured to determine a plurality of dynamic reference points at a preset position downstream of the embossing roller outlet, collect raw displacement data in real time, construct a dynamic polygon based on the dynamic reference points, calculate the centroid offset and area change rate, and generate a dynamic adjustment coefficient. The dynamic adjustment coefficient is used to calibrate the raw displacement data to obtain calibrated displacement data. A filtering module 12 is configured to perform high-frequency filtering on the calibration displacement data to generate filtered displacement data; an anti-interference module 13, configured to apply anti-electromagnetic interference processing to the filtered displacement data to generate anti-interference displacement data; A correction module 14 is configured to perform edge contour reconstruction based on a convex hull algorithm on the anti-interference displacement data and output edge reconstructed displacement data; a correction module 15 for performing temperature and humidity drift correction on the edge reconstruction displacement data to generate corrected displacement data; The alarm module 16 is used to calculate the thickness value in real time based on the corrected displacement data, and to trigger an alarm signal immediately when the thickness deviation exceeds a preset threshold and to adjust the embossing roller pressure or slitting tension synchronously.

[0020] In an embodiment of the present invention, In the thickness detection system for the specialty paper slitting and embossing process according to an embodiment of the present invention, the module 11 determines a plurality of dynamic reference points at a preset position downstream of the embossing roller outlet, collects raw displacement data in real time, constructs a dynamic polygon based on the dynamic reference points, calculates the centroid offset and area change rate, and generates a dynamic adjustment coefficient. The dynamic adjustment coefficient is used to calibrate the raw displacement data to obtain calibrated displacement data, including: Step 111, dynamically determining a plurality of dynamic reference points at a preset distance downstream of the embossing roller outlet, and collecting raw displacement data of each dynamic reference point in real time; Step 112: constructing a dynamic polygon based on the original displacement data with the dynamic reference points as vertices, and recording the spatial coordinate sequence of each vertex in real time; Step 113, calculating the coordinates of the centroid of the dynamic polygon based on the spatial coordinate sequence, comparing the coordinates of the centroid of the polygon with a preset reference centroid position to generate a centroid offset; and simultaneously calculating the area of ​​the current dynamic polygon, comparing the area of ​​the dynamic polygon with a preset reference area to generate an area change rate; Step 114: convert the centroid offset into a distance scalar value, establish a weighted function relationship between the distance scalar value and the area change rate, and output a dynamic adjustment coefficient; Step 115 , based on the dynamic adjustment coefficient, decompose the original displacement data into independent components of each reference point, and apply a compensation value matching the dynamic adjustment coefficient to each independent component; Step 116: reconstruct the compensated displacement data sequence to generate the calibration displacement data.

[0021] In an embodiment of the present invention, the selection of dynamic reference points and real-time data acquisition can adapt to the dynamic characteristics of paper movement and avoid errors of fixed reference points under complex working conditions; at the same time, the centroid offset and area change rate of the dynamic polygon can comprehensively reflect the overall deviation and local morphological changes of the paper displacement, thereby improving the comprehensiveness of the deviation analysis; the dynamic adjustment coefficient generated based on the weighted relationship can achieve differentiated compensation for different deviation factors and improve the calibration accuracy; the component compensation and data reconstruction can not only correct the displacement error of each reference point in a targeted manner, but also ensure the continuity and consistency of the calibrated data.

[0022] In the embodiments of the present invention, when applied specifically, it can be achieved through the following technical solutions, for example: In step 111 above, the location of the reference point should be selected in combination with the embossing process scenario and paper characteristics, as follows: The core selection range of the reference point is the preset distance downstream of the embossing roller outlet. The key to this position is that the paper has just left the embossing roller and its shape (such as thickness and flatness) is closest to its immediate state after embossing. It can directly reflect the impact of the embossing process on the paper thickness and avoid secondary shape changes caused by subsequent paper transmission (such as passing through other rollers and tension changes) that interfere with the reference data.

[0023] Within this range, the specific location must meet the following conditions: Distribution principle: Distribute evenly along the width of the paper (for example, select evenly from the edge to the center of the paper) to ensure that the entire horizontal state of the paper is covered to avoid affecting the judgment of the overall thickness due to local data deviation.

[0024] Avoid areas: Stay away from areas where the embossed pattern changes suddenly (such as the transition zone from dense to sparse pattern, and the area where the pattern shape changes dramatically), because the displacement changes in these areas may be caused by the morphological differences of the pattern itself rather than abnormal paper thickness. At the same time, avoid areas that are prone to interference (such as the shadow area of ​​the equipment bracket, the sensor signal blind area, the edge friction area where the paper contacts the transmission roller, etc.). The displacement measurement in these areas is easily affected by the environment or equipment, and the reliability of the original data is low.

[0025] In addition, dynamic adjustments must be made in combination with the real-time movement status and process characteristics of the paper: for example, when the paper moves at a faster speed, the reference points can be appropriately encrypted to capture rapid changes; when the tension stability is poor, priority is given to areas with smaller tension fluctuations (such as the middle of the paper, where tension fluctuations at the edges are usually greater); for special paper that is thinner and prone to wrinkling, the reference points can be biased toward the center axis of the paper where the lateral tension is more uniform, reducing the interference of wrinkles on displacement measurement; the original displacement information of these reference points during the movement of the paper is continuously monitored through displacement sensors, and the displacement values ​​of each point without any processing are continuously recorded.

[0026] In step 112, the discrete reference points are further converted into analyzable overall morphological carriers based on the original displacement data. Specifically, the preset initial position information of each dynamic reference point (such as a fixed reference position in the device coordinate system) is combined with the collected original displacement data to calculate the real-time spatial coordinates of each reference point (i.e., the physical position coordinates of the point are combined with the displacement value measured by the displacement sensor to reflect its actual spatial position). According to the actual distribution order of the reference points on the paper (such as the arrangement from the edge to the center of the paper), the real-time spatial coordinates of each point are connected in sequence as vertices to form a polygon that changes dynamically with the movement of the paper. As the paper continues to move and the displacement data is continuously updated, the system will record and update the spatial coordinate sequence of each vertex of the polygon in real time.

[0027] In step 113, the centroid position (i.e., geometric center) of the dynamic polygon is calculated based on the spatial coordinates of each vertex of the dynamic polygon using a geometric method (e.g., conventional geometric calculation rules for the centroid of a polygon). The calculated real-time centroid coordinates are compared with a preset reference centroid position (based on a standard centroid under normal production conditions) to determine the spatial offset direction and distance between the two, thereby generating a "centroid offset"—an indicator that primarily reflects the deviation in the overall position of the paper. Simultaneously, the actual area of ​​the current dynamic polygon is calculated based on the polygon vertex coordinates, compared with a preset reference area (the area of ​​the standard polygon under normal conditions), and the ratio of area change is calculated to obtain an "area change rate"—an indicator that primarily reflects the degree of deformation of the local shape of the paper.

[0028] In step 114, the centroid offset (containing vector information of direction and distance) is converted into a simple distance scalar value (retaining only the linear distance of the offset and ignoring the direction), eliminating the interference of the direction on subsequent calculations. The weight of the influence of the distance scalar value and the area change rate on the displacement calibration is analyzed (for example, the centroid offset mainly affects the overall position calibration, and the weight can be slightly higher; the area change rate mainly affects the local shape calibration, and the weight can be adjusted according to process requirements). The relationship between the two and the adjustment coefficient is established through a weighted method. Finally, a dynamic adjustment coefficient that can comprehensively reflect the degree of overall and local deviation is output, providing a specific quantitative standard for subsequent compensation.

[0029] In step 115, the displacement raw data collected in step 111 is decomposed into independent displacement components corresponding to each reference point based on the characteristics of each dynamic reference point (such as its position on the paper and its sensitivity to changes in paper thickness). This decomposes the displacement raw data into independent displacement components corresponding to each reference point (i.e., distinguishes the displacement contributions of different reference points to avoid mutual interference). Based on the dynamic adjustment coefficient output in step 114, a corresponding compensation value is matched to each independent component. Specifically, a greater compensation is applied to components with a greater degree of displacement, and a correspondingly smaller compensation is applied to components with a smaller degree of displacement, thereby achieving "compensation on demand" and ensuring the accuracy of the correction.

[0030] In step 116, the compensated independent displacement components of each reference point are reintegrated according to the temporal sequence of the paper movement (a displacement sequence that changes with time) or the spatial distribution pattern (the positional relationship along the width / length of the paper) to form a continuous and complete displacement data sequence. During the integration process, any inconsistencies between the components that may be caused by individual compensation (such as local data mutations) are eliminated, and calibrated displacement data is ultimately generated.

[0031] In the thickness detection system for the specialty paper slitting and embossing process according to an embodiment of the present invention, the module 12 performs high-frequency filtering on the calibration displacement data to generate filtered displacement data, including: Step 121: extracting high-frequency oscillation components from the displacement sequence based on the calibrated displacement data, filtering out excessive oscillation waves according to a preset vibration amplitude threshold, and generating time-domain noise-reduced displacement data; Step 122, performing frequency domain conversion processing on the time domain noise reduction displacement data to generate corresponding frequency domain energy distribution; Step 123: determining a target frequency band based on the frequency domain energy distribution and the production line operating speed, attenuating noise energy components outside the target frequency band, and generating frequency domain filtering data; Step 124 : performing time domain inverse conversion processing on the frequency domain filtered data to reconstruct it into a time domain displacement sequence to generate the filtered displacement data.

[0032] In an embodiment of the present invention, time domain noise reduction is used to directly filter out obvious high-frequency oscillation noise, reduce sudden interference in the time domain signal, and provide more stable basic data for subsequent processing; through frequency domain conversion and targeted frequency band filtering, the frequency characteristics of effective signals and noise are accurately distinguished, and the target frequency band is dynamically adjusted in combination with the production line speed to improve the adaptability and targeting of the filtering and avoid the false filtering of effective signals; through inverse conversion to reconstruct the time domain sequence, it is ensured that the processed displacement data still maintains time domain continuity, which is consistent with the time characteristics of paper movement in actual production.

[0033] In the embodiments of the present invention, when applied specifically, it can be achieved through the following technical solutions, for example: In step 121, based on the calibration displacement data, the higher-frequency oscillation components (i.e., fast, high-frequency displacement fluctuations) are separated by analyzing the fluctuation frequency characteristics of the displacement sequence. A reasonable vibration amplitude threshold is preset based on the vibration characteristics of normal paper movement in the specialty paper slitting and embossing process (oscillations exceeding this threshold are generally noise such as equipment vibration and environmental interference). The extracted high-frequency oscillation components whose amplitudes exceed this threshold are filtered out, and the displacement data that falls within the normal fluctuation range is retained, thereby generating displacement data that has undergone preliminary time-domain noise reduction.

[0034] In the above step 122, a frequency domain conversion method (such as converting the time-domain displacement sequence that changes with time into the corresponding relationship between frequency and energy) is used for the time domain denoised displacement data to convert it from a time dimension signal to a frequency dimension. Through this conversion, the energy distribution of different frequency components in the displacement data (i.e., which frequencies have strong signal energy and which have weak signal energy) is clearly presented, and the corresponding frequency domain energy distribution result is generated.

[0035] In step 123, based on the frequency domain energy distribution and in combination with the real-time operating speed of the production line (the paper movement speed will affect the frequency range of the normal displacement signal; for example, the higher the speed, the higher the characteristic frequency of the normal signal may be), a target frequency band range related to the normal movement of the paper and the embossing process (i.e., the frequency interval containing the effective displacement signal) is determined. Energy components in the frequency domain energy distribution that exceed the target frequency band range (primarily noise, such as high-frequency vibration of the equipment, electromagnetic interference, etc.) are attenuated (their energy proportion is reduced), and the effective energy within the target frequency band is retained to generate frequency domain filtered data.

[0036] In the above step 124, the frequency domain filtered data is converted from the frequency dimension back to the time dimension through a time domain inverse conversion method (corresponding to the frequency domain conversion in step 122), and a time domain displacement data sequence is reconstructed; this sequence has filtered out high-frequency noise and interference from non-target frequency bands, and ultimately generates displacement data that has undergone complete filtering processing.

[0037] In the thickness detection system for the specialty paper slitting and embossing process according to an embodiment of the present invention, the module 13 applies anti-electromagnetic interference processing to the filtered displacement data to generate anti-interference displacement data, including: Step 131: Based on the filtered displacement data, monitor the current change rate of the embossing machine power supply in real time, activate an interference event marker when the current change rate exceeds a preset mutation threshold, and record the corresponding time position; Step 132: based on the displacement data period corresponding to the interference event marker positioning, convert the displacement data of the period into a frequency domain energy spectrum; Step 133: Identify an abnormal energy peak in a preset interference frequency band in the frequency domain energy spectrum, apply a band-stop filtering process centered on the abnormal peak, suppress the abnormal peak energy to within the spectrum baseline range, and generate band-stop filtered frequency domain data; Step 134, detecting a short-duration pulse transition signal in the band-stop filtered frequency domain data, performing window average smoothing processing on the detected pulse transition, and generating corrected frequency domain data; Step 135 : Inversely convert the corrected frequency domain data into a time domain signal sequence, reconstruct it into continuous displacement data, and generate the anti-interference displacement data.

[0038] In an embodiment of the present invention, the interference marker is accurately activated by monitoring the power supply current change rate, so as to realize real-time and accurate identification of electromagnetic interference events and avoid missed detection or misjudgment; the frequency domain analysis of the interference period is focused, and the abnormal energy of the preset interference frequency band is targetedly suppressed by band-stop filtering, so as to efficiently eliminate the core frequency components of the electromagnetic interference; the residual pulse jump is processed by window average smoothing to further eliminate the local data mutation caused by interference and improve data stability; the continuous time domain data is reconstructed by inverse conversion to ensure that the processed displacement data maintains time continuity and conforms to the dynamic characteristics of actual production.

[0039] In the embodiments of the present invention, when applied specifically, it can be achieved through the following technical solutions, for example: In the above step 131, the current change of the embossing machine power supply is synchronously monitored in real time with reference to the filtered displacement data, and the rate of change of the current over time is calculated; the rate of change is compared with a preset mutation threshold (set according to the electromagnetic interference characteristics of the equipment and used to distinguish normal current fluctuations from interfering mutations). When the current change rate exceeds the threshold, the system automatically activates the electromagnetic interference event marker and accurately records the time location of the interference event (i.e., the time point corresponding to the displacement data).

[0040] In step 132, the time period corresponding to the interference event in the filtered displacement data (i.e., the interval of displacement data before and after the interference occurs) is located based on the time position recorded by the interference event marker; the time domain displacement data within the time period is converted into a frequency domain energy spectrum (by using a frequency domain conversion method to present the energy distribution of different frequency components within the time period) so as to analyze the energy characteristics of the interference from the frequency dimension.

[0041] In the above-mentioned step 133, in the frequency domain energy spectrum, based on the known frequency characteristics of the electromagnetic interference of the equipment (such as the typical interference frequency generated by the embossing machine motor, electronic control system, etc.), an abnormal energy peak (i.e., an energy protrusion that is significantly higher than the spectrum baseline) within a preset interference frequency band is identified; with the frequency corresponding to the abnormal peak as the center, a band-stop filtering process is applied (targetedly suppressing the energy of this frequency and nearby frequency bands) to suppress the energy of the abnormal peak to within the fluctuation range of the spectrum baseline, thereby eliminating the electromagnetic interference energy in this frequency band and generating frequency domain data after band-stop filtering.

[0042] In step 134 , the band-stop filtered frequency domain data is detected for short-duration pulse transition signals (i.e., abnormal signals with short duration and sudden amplitude changes) remaining due to electromagnetic interference. For each detected pulse transition, a certain time window is defined with the transition point as the center (covering the continuous data before and after the transition), and the average value of the data within the window is calculated. The average value is used to replace the abnormal value of the transition point to achieve smoothing processing, and finally generate the corrected frequency domain data.

[0043] In the above step 135, the corrected frequency domain data is converted from the frequency dimension back to the time dimension through a time domain inverse conversion method (corresponding to the frequency domain conversion in step 132), and a continuous time domain displacement data sequence is reconstructed; the continuity of the data on the time axis is ensured through integration processing, and the breaks or mutations that may occur during the interference processing are eliminated, and finally displacement data that is resistant to electromagnetic interference is generated.

[0044] In the thickness detection system for the specialty paper slitting and embossing process according to an embodiment of the present invention, the module 14 performs edge contour reconstruction based on a convex hull algorithm on the anti-interference displacement data and outputs the edge reconstructed displacement data, including: Step 141 , establishing a two-dimensional structured grid coordinate system based on the anti-interference displacement data, wherein the coordinate system takes the center of the sensor measurement spot as the origin, and determines the grid unit size according to a preset spot size parameter and a burr influence range; Step 142, extracting spatial distribution data of burr fibers within a cell in the two-dimensional structured grid coordinate system, constructing a convex hull boundary of the burr region based on the spatial distribution data, and calculating a light scattering attenuation weight factor based on the distance between the convex hull boundary and the cell center; Step 143, processing the attenuation weight factors of all grid cells to generate a grid cell attenuation coefficient matrix; Step 144 , mapping the anti-interference displacement data to a grid coordinate system, determining for each displacement data point the grid unit to which it belongs, calculating a displacement compensation value based on the attenuation coefficient of the unit, and correcting the original displacement data using the compensation value to obtain compensated displacement data; Step 145 : Verify the edge area of ​​the compensated displacement data, reconstruct a continuous displacement sequence based on the verified displacement data, and output the edge reconstructed displacement data.

[0045] In an embodiment of the present invention, a grid coordinate system matching the light spot and burrs is established to provide a spatial reference for fine analysis of the burr effect, thereby ensuring the spatial accuracy of subsequent processing. A convex hull algorithm is used to construct the burr boundary, accurately capture the actual contour of the burr, and combine the distance to calculate the attenuation weight factor to achieve quantification of light scattering interference and effectively eliminate the measurement deviation caused by the burrs. An attenuation coefficient matrix is ​​used to systematize the scattered burr interference information, providing an efficient tool for batch compensation of displacement data. Through grid mapping and coefficient compensation, the light scattering error of each data point is corrected in a targeted manner to improve the absolute accuracy of the displacement data. Verification and reconstruction of the edge area ensure the reliability and continuity of the edge data and avoid edge contour distortion caused by burr interference.

[0046] In the embodiments of the present invention, when applied specifically, it can be achieved through the following technical solutions, for example: In step 141 , a two-dimensional structured grid coordinate system is established based on the anti-interference displacement data. The origin of the coordinate system is precisely located at the center of the displacement sensor's measurement spot (i.e., the focal point of the measurement beam emitted by the sensor), ensuring that all measurement data can be spatially positioned relative to the center of the spot.

[0047] The size of the grid cell is a key parameter, determined by considering two core factors: the preset spot size parameters (such as the diameter of the sensor spot and the actual spatial coverage), and the typical impact range of the burrs on specialty paper (i.e., the area where burrs may scatter the sensor's light). Specifically, the grid cell size must be smaller than the minimum resolution of the spot (to ensure that subtle changes within the spot can be captured) and the minimum scale of burr impact (to avoid missing the effects of small burrs). For example, if the minimum spot resolution is 0.1mm and the minimum burr impact scale is 0.05mm, the cell size can be set to less than 0.05mm. This design of the grid not only fully covers the sensor's measurement range (ensuring that all measurement points are within the grid), but also accurately reflects the spatial distribution details of the burr fibers, laying the foundation for subsequent accurate analysis of burr interference.

[0048] In the above step 142, each grid unit is analyzed one by one in the constructed grid coordinate system to deeply capture the interference pattern of burrs on light. Specifically, the spatial distribution data of burr fibers in each unit is extracted, including the number of burr fibers in the unit, the specific position coordinates of each fiber (X, Y coordinates in the grid), the distribution density (the number of fibers per unit area), etc., to fully understand the distribution status of burrs in the unit.

[0049] Based on these distribution data, the convex hull algorithm is used to construct the convex hull boundary of the burr area within the unit: the convex hull boundary is the minimum convex polygon that can completely enclose all the burr fibers in the unit, and its shape directly reflects the actual contour range of the burr fibers (for example, when the burr fibers are dispersed, the convex hull boundary will be more "loose"; when the fibers are concentrated, the boundary will be more "compact").

[0050] The straight-line distance from the convex hull boundary to the cell center is calculated: the closer the distance, the closer the burr fibers are to the cell center (i.e., the core measurement area of ​​the sensor spot), and the greater the light scattering interference; the farther the distance, the weaker the scattering interference. Based on the correlation between this distance and scattering intensity, a light scattering attenuation weight factor is generated for that grid cell. This factor quantifies the degree of light scattering interference; a larger weight indicates a greater measurement error caused by the burr fibers in that cell.

[0051] In step 143 , the light scattering attenuation weight factors of all grid cells are summarized and organized into a structured grid cell attenuation coefficient matrix according to the arrangement order of the grid cells in the two-dimensional coordinate system (e.g., first by row and then by column).

[0052] Each element in the matrix corresponds to the attenuation coefficient of a grid cell, and the number of rows and columns of the matrix is ​​consistent with the number of rows and columns of the grid. For example, if the grid is 100 rows × 100 columns, the matrix is ​​a 100 × 100 two-dimensional array, and each element (i, j) corresponds to the attenuation coefficient of the grid cell in the i-th row and j-th column. This matrix intuitively presents the spatial distribution of light scattering interference in the entire measurement range (which areas have strong attenuation and which areas have weak attenuation), providing a unified quantitative basis for subsequent rapid query of the interference degree of each displacement data point and batch calculation of compensation values.

[0053] In step 144 , the anti-interference displacement data (which has been freed from electromagnetic interference and other influences, but may still be affected by light scattering from burrs) is mapped to the grid coordinate system established in step 141 . By matching coordinates (i.e., the spatial coordinates of the displacement data points correspond to the grid coordinates), it is determined to which grid cell each displacement data point belongs.

[0054] The attenuation coefficient corresponding to the grid cell in the attenuation coefficient matrix generated in step 143 is used to calculate the displacement compensation value based on the original displacement data for that point. For example, if the attenuation coefficient is 0.8 (indicating strong scattering interference), and the original displacement measurement is 0.05 mm smaller due to scattering, the compensation value may be +0.05 mm. If the attenuation coefficient is 0.2 (indicating weak interference), and the measurement is 0.01 mm off, the compensation value may be +0.01 mm. Finally, the calculated compensation value is used to correct the original displacement data for that data point, generating displacement data that has been compensated for light scattering errors.

[0055] In step 145 , the compensated displacement data is primarily verified for the edge area of ​​the paper. This is where burrs are most concentrated and measurement errors are most likely to accumulate. Therefore, verification includes the continuity of the edge data (checking for irregular mutations, such as sudden increases or decreases in displacement values ​​of adjacent points) and its coordination with the data in the normal area (determining whether the changes in the edge data conform to the natural variation in paper thickness from the center to the edge, such as whether there is a disconnect with the normal area due to overcompensation).

[0056] Verified displacement data (i.e., continuous and coordinated data) are reintegrated according to the time sequence of paper movement (following the order of the production process) or spatial distribution pattern (position along the width / length direction of the paper), eliminating possible local data breaks and reconstructing a complete and continuous displacement data sequence; the edge reconstruction displacement data finally generated has its edge contour accurately corrected and can truly reflect the actual shape of the paper edge.

[0057] In the thickness detection system for the specialty paper slitting and embossing process according to an embodiment of the present invention, the module 15 performs temperature and humidity drift correction on the edge reconstruction displacement data to generate corrected displacement data, including: Step 151: collecting ambient temperature sensor data and ambient humidity sensor data in real time; Step 152 , calculating the difference between the current temperature and the reference temperature based on the temperature sensor data, and generating an original linear compensation component according to a preset temperature compensation coefficient; Step 153: Calculate the difference between the current humidity and the reference humidity based on the humidity sensor data, and generate an original nonlinear compensation component according to a preset humidity compensation coefficient; Step 154: performing normalization processing on the original linear compensation component and the original nonlinear compensation component respectively to obtain normalized linear compensation component and nonlinear compensation component; Step 155 , generating a basic composite value through weighted fusion calculation based on the normalized linear compensation component and the nonlinear compensation component; Step 156, generating a final composite compensation value based on the basic composite value and the time delay parameter compensation; Step 157 , dynamically superimposing the final composite compensation value onto the edge reconstructed displacement data in a time series to generate preliminary corrected displacement data; Step 158 : verifying the stability of the displacement data under constant temperature and humidity conditions, and generating the corrected displacement data when the drift amount is lower than a preset drift tolerance threshold.

[0058] In an embodiment of the present invention, by separating the linear influence of temperature from the nonlinear influence of humidity, compensation components are generated in a targeted manner, thereby improving the accuracy of drift correction; normalization processing eliminates the dimensional differences of the temperature and humidity compensation components, providing a unified basis for subsequent fusion; through weighted fusion and time delay compensation, the weight differences and hysteresis characteristics of the temperature and humidity influences are taken into account, making the compensation more in line with actual physical laws; dynamic superposition ensures the time synchronization of compensation and displacement data, avoiding deviations caused by static compensation; constant temperature and humidity verification confirms data stability through actual measurement, ensuring that the correction results meet the process accuracy requirements.

[0059] In the embodiments of the present invention, when applied specifically, it can be achieved through the following technical solutions, for example: In step 151, the ambient temperature sensor and humidity sensor deployed near the detection system continuously collect the temperature data (e.g., recording the temperature value every 0.1 second) and humidity data (e.g., synchronously recording the humidity value) of the current production environment in real time, and mark each piece of data with a corresponding timestamp to ensure the temporal correlation between the temperature and humidity data and the displacement data.

[0060] In step 152, a preset reference temperature (e.g., the standard operating temperature for the specialty paper slitting and embossing process, e.g., 25°C) is used as a reference, and the reference temperature is subtracted from the current temperature value to obtain a temperature difference (i.e., temperature fluctuation). Based on a preset temperature compensation coefficient (this coefficient is determined experimentally and reflects the linear drift of the displacement measurement value for every 1°C change in temperature) of the system, the temperature difference is combined with the temperature compensation coefficient to calculate an original linear compensation component (this component exhibits a linear relationship with temperature) for offsetting the temperature effect.

[0061] In step 153, a preset reference humidity (e.g., process standard humidity, such as 50% RH) is used as a reference, and the reference humidity is subtracted from the current humidity value collected in step 151 to obtain a humidity difference (i.e., humidity fluctuation). Based on a preset humidity compensation coefficient (this coefficient is determined based on the nonlinear effect of humidity on paper expansion / contraction, e.g., the effect is more significant at high humidity), the humidity difference is combined with the humidity compensation coefficient to calculate an original nonlinear compensation component for offsetting the humidity effect (this component varies nonlinearly with humidity, e.g., the higher the humidity, the faster the compensation amount increases).

[0062] In step 154, the original linear compensation component generated in step 152 and the original nonlinear compensation component generated in step 153 are normalized respectively. That is, by adjusting the numerical ranges of the two components, they are uniformly mapped to the same standard interval (such as 0-1). For example, if the original range of the linear component is 0-2 mm and the nonlinear component is 0-5 mm, they are both adjusted to 0-1 after normalization, eliminating the fusion deviation caused by the difference in dimension or numerical range between the two.

[0063] In step 155, the normalized linear compensation component and the nonlinear compensation component are weightedly fused according to the actual influence weights of temperature and humidity on displacement drift (e.g., a higher weight is given when the temperature influence is greater, which can be set based on historical process data). For example, if the temperature weight is 0.6 and the humidity weight is 0.4, the linear component is multiplied by 0.6 and the nonlinear component is multiplied by 0.4, and the two results are then added together to generate a basic composite value that comprehensively reflects the combined influence of temperature and humidity.

[0064] In step 156 , a preset time delay parameter (based on a typical experimentally measured lag time) is introduced to compensate for the time dimension of the basic composite value, taking into account the time delay caused by the impact of temperature and humidity changes on displacement (e.g., when the ambient humidity increases, the displacement change may lag by 10 seconds due to moisture absorption and expansion of the paper). For example, if the delay parameter is 10 seconds, the current basic composite value is correlated with the composite value 10 seconds ago, ultimately generating a final composite compensation value that better reflects the actual impact pattern.

[0065] In step 157, the final composite compensation value is dynamically matched with the edge reconstructed displacement data generated by module 14 according to a time series. That is, the displacement data at each time point is superimposed with the final composite compensation value at that time point (e.g., the displacement data at time t is superimposed with the compensation value at time t). This dynamic superposition eliminates the measurement deviation caused by temperature and humidity drift and generates preliminary corrected displacement data.

[0066] In step 158, the preliminary corrected displacement data is placed in a constant temperature and humidity environment (i.e., an environment where the temperature and humidity are stable at the reference values) for verification, and its data stability is continuously monitored (e.g., the displacement fluctuation within 10 consecutive minutes). When the monitored drift (i.e., the deviation between the actual value and the theoretical standard value) is lower than the preset drift tolerance threshold (e.g., 0.01 mm), it indicates that the compensation effect meets the standard, and finally, displacement data that has been corrected for temperature and humidity drift is generated.

[0067] In the thickness detection system for the specialty paper slitting and embossing process according to an embodiment of the present invention, the module 16 calculates the thickness value in real time based on the corrected displacement data, and when the thickness deviation exceeds a preset threshold, immediately triggers an alarm signal and synchronously adjusts the embossing roller pressure or slitting tension, including: Step 161 , separating the upper and lower surface displacement components based on the corrected displacement data, and calculating the displacement difference at corresponding time points to generate a real-time thickness sequence; Step 162 , comparing the real-time thickness sequence with the preset standard thickness value, and dynamically calculating the thickness deviation value; Step 163: When the thickness deviation value exceeds the preset deviation tolerance, an over-limit alarm instruction and deviation characteristic parameters are generated; Step 164, executing a control decision based on the deviation characteristic parameter, specifically includes: The direction of the actuator is determined according to the positive or negative sign of the deviation value, and a single-mechanism or dual-mechanism coordinated adjustment strategy is selected according to the deviation amplitude to generate an instruction group for the embossing roller pressure adjustment amount and the slitting tension adjustment amount; Step 165 , immediately triggering a multi-modal alarm signal, and synchronously sending the command group to the actuator via a preset industrial communication protocol; Step 166 , obtaining the actuator feedback signal in real time, and completing the closed-loop calibration when it is detected that the adjusted secondary thickness value returns to the standard range.

[0068] In an embodiment of the present invention, the thickness is calculated by separating the upper and lower surface displacement components, and the physical relationship between displacement and thickness is directly linked to ensure the accuracy of thickness calculation; dynamic monitoring and timely alarm of thickness deviation are achieved to avoid accumulation of thickness anomalies and defects in batch products; accurate decision-making and adjustment strategies are made based on deviation characteristic parameters, taking into account adjustment direction, amplitude and mechanism coordination to improve the targeted control; multi-modal alarm and industrial protocol transmission are used to ensure timely response of operators and rapid action of actuators, thereby shortening the abnormality handling time; closed-loop control is achieved through feedback signals and secondary thickness monitoring to ensure that the adjustment effect meets the standards and avoid blind adjustment.

[0069] In the embodiments of the present invention, when applied specifically, it can be achieved through the following technical solutions, for example: In step 161 , based on the corrected displacement data, the displacement components of the upper and lower surfaces of the paper are extracted by separating the data features (e.g., based on the phase difference and amplitude characteristics of the displacement signals of the upper and lower surfaces). At the same time point, the difference between the upper and lower surface displacement components (i.e., the distance between the upper and lower surfaces) is calculated, and the differences between consecutive time points are arranged in sequence to generate a sequence of data reflecting the real-time thickness changes of the paper (a real-time thickness sequence).

[0070] In step 162, the thickness value at each time point in the real-time thickness sequence is compared with a preset standard thickness value (a target thickness set according to the process requirements of specialty paper, such as 0.1 mm ± 0.005 mm). The thickness deviation value at each time point is dynamically calculated by subtracting the standard thickness value from the real-time thickness value, forming a dynamic record of the deviation value changing over time (e.g., a positive deviation indicates that the thickness is too thick, and a negative deviation indicates that the thickness is too thin).

[0071] In step 163, the dynamically calculated thickness deviation value is compared in real time with a preset deviation tolerance (the maximum deviation range allowed by the process, such as ±0.01 mm). When the deviation value exceeds the tolerance (e.g., greater than 0.01 mm or less than -0.01 mm for three consecutive sampling cycles), the system automatically generates an over-limit alarm instruction (a control signal that triggers the alarm) and simultaneously extracts deviation characteristic parameters (including the specific value of the deviation, duration, and distribution position in the paper width direction, etc., for subsequent control decisions).

[0072] In step 164, control decisions are made based on the deviation characteristic parameters: The direction of action of the actuator is determined according to the positive or negative sign of the deviation value (for example, a positive deviation (too thick thickness) may require increasing the pressure of the embossing roller to thin the paper, or reducing the slitting tension to avoid excessive stretching; a negative deviation (too thin thickness) is the opposite); Select the adjustment strategy based on the deviation amplitude (the absolute value of the deviation): if the deviation is small (e.g. less than 0.02mm), use single-mechanism adjustment (adjust only the embossing roller pressure or only the slitting tension); if the deviation is large (e.g. greater than or equal to 0.02mm), use dual-mechanism coordinated adjustment (adjust pressure and tension simultaneously); Finally, a specific instruction group of the embossing roller pressure adjustment amount (such as increasing / decreasing the pressure value) and the slitting tension adjustment amount (such as increasing / decreasing the tension value) is generated according to the above decision.

[0073] In step 165, the system immediately triggers a multi-modal alarm signal (e.g., an audible and visual alarm to alert the on-site operator, a text alarm message displayed on the control system screen, and an alarm notification pushed to the remote terminal). Simultaneously, the pressure and tension adjustment instruction set is synchronously sent to the corresponding actuators (e.g., the hydraulic servo system for the embossing roller, the magnetic powder brake for the slitting tension, etc.) via a preset industrial communication protocol (e.g., Modbus, CAN bus), ensuring rapid transmission of the adjustment instructions.

[0074] In step 166, the feedback signal returned by the actuator (such as the actual pressure adjustment value of the hydraulic servo system and the actual tension output value of the magnetic powder brake) is received in real time to confirm whether the instruction is executed correctly. At the same time, the adjusted paper thickness (i.e., the secondary thickness value) is continuously monitored through real-time thickness calculation logic. When it is detected that the secondary thickness value returns to the preset standard thickness range (and is stable for a period of time, such as no over-limit for five consecutive sampling periods), the adjustment is determined to be effective, and the closed-loop correction of the thickness anomaly is completed.

[0075] like Figure 2 As shown, a control method for a thickness detection system for a specialty paper slitting and embossing process is provided, the control method comprising the following steps: Determine multiple dynamic reference points at a preset position downstream of the embossing roller outlet, collect original displacement data in real time, construct dynamic polygons based on the dynamic reference points, calculate the centroid offset and area change rate, and generate dynamic adjustment coefficients. Use the dynamic adjustment coefficients to calibrate the original displacement data to obtain calibrated displacement data. performing high-frequency filtering on the calibration displacement data to generate filtered displacement data; applying anti-electromagnetic interference processing to the filtered displacement data to generate anti-interference displacement data; Performing edge contour reconstruction based on a convex hull algorithm on the anti-interference displacement data, and outputting edge reconstructed displacement data; performing temperature and humidity drift correction on the edge reconstructed displacement data to generate corrected displacement data; The thickness value is calculated in real time based on the corrected displacement data, and when the thickness deviation exceeds a preset threshold, an alarm signal is immediately triggered and the embossing roller pressure or slitting tension is adjusted synchronously.

[0076] An embodiment of the present invention further provides a computing device comprising: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the above-described method. All implementations in the above-described method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0077] The embodiment of the present invention further provides a computer-readable storage medium storing instructions, which, when executed on a computer, causes the computer to execute the above-described method. All implementations in the above-described method embodiment are applicable to this embodiment and can achieve the same technical effects.

[0078] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some or all of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A thickness detection system for special paper slitting and embossing process, characterized by: include: An acquisition module is used to determine multiple dynamic reference points at a preset position downstream of the embossing roller outlet, collect raw displacement data in real time, construct dynamic polygons based on the dynamic reference points, calculate the centroid offset and area change rate, and generate dynamic adjustment coefficients. The dynamic adjustment coefficients are used to calibrate the raw displacement data to obtain calibrated displacement data; A filtering module, configured to perform high-frequency filtering on the calibration displacement data to generate filtered displacement data; an anti-interference module, configured to apply anti-electromagnetic interference processing to the filtered displacement data to generate anti-interference displacement data; a correction module, configured to perform edge contour reconstruction based on a convex hull algorithm on the anti-interference displacement data and output edge reconstructed displacement data; a correction module, configured to perform temperature and humidity drift correction on the edge reconstruction displacement data to generate corrected displacement data; The alarm module is used to calculate the thickness value in real time based on the corrected displacement data, and to trigger an alarm signal immediately and synchronously adjust the embossing roller pressure or slitting tension when the thickness deviation exceeds a preset threshold.

2. The thickness detection system for specialty paper slitting and embossing process according to claim 1 is characterized in that: Determine multiple dynamic reference points at a preset position downstream of the embossing roller outlet, collect original displacement data in real time, construct dynamic polygons based on the dynamic reference points, calculate the centroid offset and area change rate, and generate dynamic adjustment coefficients. Use the dynamic adjustment coefficients to calibrate the original displacement data to obtain calibrated displacement data, including: Dynamically determine multiple dynamic reference points at a preset distance downstream of the embossing roller outlet, and collect the original displacement data of each dynamic reference point in real time; Based on the displacement original data, a dynamic polygon is constructed with the dynamic reference point as a vertex, and the spatial coordinate sequence of each vertex is recorded in real time; Calculating the centroid coordinates of the dynamic polygon based on the spatial coordinate sequence, and comparing the centroid coordinates of the polygon with a preset reference centroid position to generate a centroid offset; and simultaneously calculating the area of ​​the current dynamic polygon, and comparing the area of ​​the dynamic polygon with a preset reference area to generate an area change rate; Converting the centroid offset into a distance scalar value, establishing a weighted function relationship between the distance scalar value and the area change rate, and outputting a dynamic adjustment coefficient; Based on the dynamic adjustment coefficient, the original displacement data is decomposed into independent components of each reference point, and a compensation value matching the dynamic adjustment coefficient is applied to each independent component; The compensated displacement data sequence is reconstructed to generate the calibration displacement data.

3. The thickness detection system for specialty paper slitting and embossing process according to claim 2, characterized in that: Performing high-frequency filtering on the calibration displacement data to generate filtered displacement data includes: Based on the calibrated displacement data, high-frequency oscillation components in the displacement sequence are extracted, and excessive oscillation waves are filtered out according to a preset vibration amplitude threshold to generate time-domain noise-reduced displacement data; Performing frequency domain conversion processing on the time domain noise reduction displacement data to generate corresponding frequency domain energy distribution; Based on the frequency domain energy distribution and in combination with the production line running speed, a target frequency band range is determined, noise energy components outside the target frequency band range are attenuated, and frequency domain filtering data is generated; The frequency domain filtered data is subjected to time domain inverse conversion processing to be reconstructed into a time domain displacement sequence to generate the filtered displacement data.

4. The thickness detection system for specialty paper slitting and embossing process according to claim 3 is characterized in that: Applying anti-electromagnetic interference processing to the filtered displacement data to generate anti-interference displacement data, including: Based on the filtered displacement data, the current change rate of the embossing machine power supply is monitored in real time, and when the current change rate exceeds a preset mutation threshold, an interference event marker is activated and the corresponding time position is recorded; Based on the displacement data period corresponding to the interference event marker positioning, converting the displacement data of the period into a frequency domain energy spectrum; Identifying an abnormal energy peak in a preset interference frequency band in the frequency domain energy spectrum, applying a band-stop filtering process centered on the abnormal peak to suppress the abnormal peak energy within the spectrum baseline range, and generating band-stop filtered frequency domain data; Detecting a short-duration pulse jump signal in the band-stop filtered frequency domain data, performing window average smoothing processing on the detected pulse jump, and generating corrected frequency domain data; The corrected frequency domain data is inversely converted into a time domain signal sequence, reconstructed into continuous displacement data, and the anti-interference displacement data is generated.

5. The thickness detection system for specialty paper slitting and embossing process according to claim 4 is characterized in that: Performing edge contour reconstruction based on a convex hull algorithm on the anti-interference displacement data and outputting edge reconstructed displacement data, including: A two-dimensional structured grid coordinate system is established based on the anti-interference displacement data, wherein the coordinate system takes the center of the sensor measurement spot as the origin, and the grid unit size is determined according to the preset spot size parameters and the burr influence range; In the two-dimensional structured grid coordinate system, extracting spatial distribution data of burr fibers within a unit, constructing a convex hull boundary of the burr region based on the spatial distribution data, and calculating a light scattering attenuation weight factor according to the distance between the convex hull boundary and the unit center; Process the attenuation weight factors of all grid cells to generate a grid cell attenuation coefficient matrix; Mapping the anti-interference displacement data to a grid coordinate system, determining the grid unit to which each displacement data point belongs, calculating the displacement compensation value based on the attenuation coefficient of the unit, and correcting the original displacement data using the compensation value to obtain the compensated displacement data; Verification is performed on the edge area of ​​the compensated displacement data, a continuous displacement sequence is reconstructed based on the verified displacement data, and the edge reconstructed displacement data is output.

6. The thickness detection system for specialty paper slitting and embossing process according to claim 5, characterized in that: Performing temperature and humidity drift correction on the edge reconstructed displacement data to generate corrected displacement data includes: Real-time collection of ambient temperature sensor data and ambient humidity sensor data; Calculate the difference between the current temperature and the reference temperature based on the temperature sensor data, and generate the original linear compensation component according to the preset temperature compensation coefficient; Calculate the difference between the current humidity and the reference humidity based on the humidity sensor data, and generate the original nonlinear compensation component according to the preset humidity compensation coefficient; Normalizing the original linear compensation component and the original nonlinear compensation component to obtain normalized linear compensation component and nonlinear compensation component; Based on the normalized linear compensation component and nonlinear compensation component, a basic composite value is generated through weighted fusion calculation; Generating a final composite compensation value according to the basic composite value and the time delay parameter compensation; Dynamically superimposing the final composite compensation value onto the edge reconstructed displacement data in a time series to generate preliminary corrected displacement data; The stability of the displacement data is verified under constant temperature and humidity conditions, and when the drift amount is lower than a preset drift tolerance threshold, the corrected displacement data is generated.

7. The thickness detection system for specialty paper slitting and embossing process according to claim 6, characterized in that: The thickness value is calculated in real time based on the corrected displacement data. When the thickness deviation exceeds a preset threshold, an alarm signal is immediately triggered and the embossing roller pressure or slitting tension is adjusted synchronously, including: Separating the upper and lower surface displacement components based on the corrected displacement data, and calculating the displacement differences at corresponding time points to generate a real-time thickness sequence; Compare the real-time thickness sequence with the preset standard thickness value and dynamically calculate the thickness deviation value; When the thickness deviation value exceeds the preset deviation tolerance, an over-limit alarm instruction and deviation characteristic parameters are generated; Executing a control decision based on the deviation characteristic parameter specifically includes: The direction of the actuator is determined according to the positive or negative sign of the deviation value, and a single-mechanism or dual-mechanism coordinated adjustment strategy is selected according to the deviation amplitude to generate an instruction group for the embossing roller pressure adjustment amount and the slitting tension adjustment amount; Instantly trigger a multi-modal alarm signal and synchronously send the command group to the actuator through a preset industrial communication protocol; The actuator feedback signal is obtained in real time, and when it is detected that the adjusted secondary thickness value returns to the standard range, the closed-loop correction is completed.

8. A method for detecting thickness of specialty paper during slitting and embossing, characterized in that: Applied to the system according to any one of claims 1 to 7, the method comprises: Determine multiple dynamic reference points at a preset position downstream of the embossing roller outlet, collect original displacement data in real time, construct dynamic polygons based on the dynamic reference points, calculate the centroid offset and area change rate, and generate dynamic adjustment coefficients. Use the dynamic adjustment coefficients to calibrate the original displacement data to obtain calibrated displacement data. performing high-frequency filtering on the calibration displacement data to generate filtered displacement data; applying anti-electromagnetic interference processing to the filtered displacement data to generate anti-interference displacement data; Performing edge contour reconstruction based on a convex hull algorithm on the anti-interference displacement data, and outputting edge reconstructed displacement data; performing temperature and humidity drift correction on the edge reconstructed displacement data to generate corrected displacement data; The thickness value is calculated in real time based on the corrected displacement data, and when the thickness deviation exceeds a preset threshold, an alarm signal is immediately triggered and the embossing roller pressure or slitting tension is adjusted synchronously.

9. A computing device, characterized in that include: one or more processors; A storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the system according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a program, which, when executed by a processor, implements the system according to any one of claims 1 to 7.

Images