Intelligent detection method for UV ink curing degree of offset printing machine

Abstract

The present application belongs to the technical field of printing quality intelligent detection, and relates to a UV ink curing degree intelligent detection method of a offset printing machine. The present application determines the area code through the conveying encoder pulse counting and the excitation light spot size, acquires the dot area rate and the first fluorescent response value before curing, collects the second fluorescent response value, the ink layer thickness, the optical coefficient of the excitation and emission wave band after curing, calculates the fluorescent response change value, inputs the dot area rate, the ink layer thickness and the optical coefficient into the photon transport probability model to obtain the equivalent optical path, constructs the optical path attenuation factor to compensate the fluorescent response change value to obtain the target curing characteristic value, and outputs the curing rate prediction value. The present application solves the problems of time lag, sampling limitation, ink layer structure interference and poor position matching in the existing detection, realizes real-time acquisition of the real curing degree of each area, eliminates the influence of non-curing factors, and improves the reliability and accuracy of online detection.

Description

Technical Field

[0001] This invention belongs to the field of intelligent printing quality detection technology, and relates to an intelligent detection method for the curing degree of UV ink in offset printing machines. Background Technology

[0002] When offset printing machines use UV inks, the ink forms a cured film layer after being irradiated by ultraviolet light. The degree of curing directly affects the dryness, abrasion resistance, and adhesion of the printed surface, as well as the stable operation of subsequent processes such as stacking, lamination, and die-cutting. During production, changes in printing speed, UV irradiation energy, ink coverage, substrate surface characteristics, and environmental conditions can all cause differences in the degree of UV ink curing at different printing locations. Therefore, online monitoring of the UV ink curing degree during offset printing machine operation is a crucial step in ensuring printing quality and process control stability.

[0003] Current methods for testing the curing of UV inks mostly employ manual sampling, surface wiping, tape peeling, or offline instrument testing. These methods are typically performed after the printed product is output, resulting in a time lag between the test results and the actual printing process. The results primarily reflect the curing state of the sampled location. Under continuous high-speed operation of offset printing presses, differences in image areas, ink coverage, and the surface condition of the substrate can cause variations in test results, affecting the accuracy of online judgment of curing degree.

[0004] Some online optical inspection methods characterize the curing degree of UV ink by collecting changes in the optical response of the ink after exposure to light. However, in offset printing, the detection signal is simultaneously affected by factors such as image depth, ink coverage ratio, ink layer morphology, substrate reflectivity, and the correspondence between detection positions. When these factors change, the optical response change may include components other than curing changes, leading to deviations in the intelligent detection results of the UV ink curing degree on the offset printing machine.

[0005] Furthermore, offset printing presses are continuously transported before and after UV irradiation, and the accuracy of the detection position will affect the comparability of the detection data before and after irradiation. After UV irradiation, the film structure of the ink and the propagation state of light at the ink-substrate interface will also change, thus affecting the correspondence between the optical detection signal and the actual degree of curing. Summary of the Invention

[0006] In view of this, in order to solve the problems mentioned in the background technology, an intelligent detection method for the curing degree of UV ink in offset printing machines is proposed.

[0007] The objective of this invention can be achieved through the following technical solution: an intelligent detection method for the curing degree of UV ink in an offset printing machine, comprising: S1, determining the area code of the printing area to be tested based on the pulse count of the offset printing machine's conveyor encoder and the size of the excitation spot.

[0008] S2. Before UV curing, obtain the dot area ratio of the printing area to be tested corresponding to the area code, and collect the first fluorescence response value.

[0009] S3. After UV curing, according to the area code, match the same printing area to be tested, collect the second fluorescence response value of the same type as the first fluorescence response value, and obtain the ink layer thickness after curing and the absorption coefficient and scattering coefficient of the ink under the excitation light band and the emission light band.

[0010] S4. Calculate the difference between the first fluorescence response value and the second fluorescence response value as the fluorescence response change value; input the dot area ratio, the thickness of the ink layer after curing, and the absorption coefficient and scattering coefficient under the excitation light band and the emission light band into the photon transport probability model, and output the equivalent optical path; calculate the ratio of the equivalent optical path to the reference optical path as the optical path attenuation factor, and divide the fluorescence response change value by the optical path attenuation factor to obtain the target curing characteristic value.

[0011] S5. Input the target solidification feature value into the solidification degree identification model, and output the solidification rate prediction value corresponding to the region code.

[0012] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The present invention solves the problem that the existing methods of manual sampling, surface wiping or offline instrument testing have time intervals and only reflect the sampling position status, which affects the accuracy of online judgment when running at high speed. It achieves the effect of obtaining the real curing degree of each area in real time during the operation of the offset printing machine, eliminating sampling limitations and improving the accuracy of online judgment.

[0013] (2) The present invention effectively solves the problem of interference from non-curing factors such as mixed images and ink layers in the prior art by inputting parameters reflecting the physical structure and optical properties of ink into the photon transport probability model, calculating the optical path attenuation factor, and compensating for the change value of fluorescence response. It achieves the effect of removing the influence of physical structure from the fluorescence signal and obtaining the target curing characteristic value that is only highly related to the degree of curing.

[0014] (3) The present invention solves the problem of poor detection position correspondence accuracy of printed products under continuous conveying state, which affects the comparability of detection data before and after the test by determining the area code of the printing area to be tested according to the pulse count and the size of the excitation spot, and matching the same printing area to be tested according to the area code before and after ultraviolet irradiation. It achieves the effect of accurate positioning and matching of the same spatial interval before and after ultraviolet irradiation, ensuring that the fluorescence response values ​​before and after originate from the same area, and improving the curing degree and detection reliability. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a flowchart of an intelligent detection method for the curing degree of UV ink in an offset printing machine according to the present invention;

[0017] Figure 2 This is a thermodynamic distribution diagram of the optical path attenuation factor in this invention;

[0018] Figure 3 This is a fitting verification diagram of the target curing feature value and the predicted curing rate value in this invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0021] The following describes in detail, with reference to the accompanying drawings, a specific scheme for an intelligent detection method for the curing degree of UV ink in an offset printing machine provided by the present invention.

[0022] Please see Figure 1 As shown, the implementation of the present invention includes S1 to S5: S1, determining the area code of the printing area to be tested based on the pulse count of the offset printing machine's conveyor encoder and the size of the excitation spot.

[0023] Considering that the printing area to be tested moves continuously along the conveying direction with the substrate during the operation of the offset printing machine, it is difficult to directly locate the printing area to be tested using fixed coordinates. The size of the excitation spot formed by the detection probe in the conveying direction determines the spatial range that can be distinguished during detection; if the area division is too dense, the detection signals of adjacent areas are prone to overlap; if the area division is too sparse, the data collected before and after UV curing may not accurately correspond to the same printing area to be tested.

[0024] Therefore, this step uses the pulse signal output by the offset printing machine's conveyor encoder to characterize the conveying displacement of the substrate, and combines the size of the excitation spot in the conveying direction to divide the substrate into continuous intervals, generating a region code corresponding to the printing area to be tested, thereby providing a positional basis for matching the same printing area to be tested before and after UV curing.

[0025] In one specific embodiment, firstly, the offset printing machine's conveyor encoder synchronously outputs pulse signals as the substrate conveying mechanism moves. The data acquisition module receives and counts these pulse signals. The encoder resolution represents the substrate conveying displacement corresponding to a single pulse, which can be determined according to the factory specifications of the offset printing machine's conveyor encoder. Multiplying the pulse count by the encoder resolution yields the substrate displacement along the conveying direction.

[0026] Subsequently, the full width at half maximum (FWHM) of the excitation spot in the transport direction during fluorescence signal acquisition is set as the region division step size. Specifically, the FWHM is determined by measuring a beam analyzer. This involves acquiring the two-dimensional light intensity distribution of the excitation spot at the working distance of the detection probe, selecting a cross-section passing through the light intensity peak and parallel to the substrate transport direction as the measurement cross-section, determining two positions on this cross-section where the light intensity is equal to half the peak value, and using the distance between these two positions as the FWHM of the excitation spot in the transport direction.

[0027] Finally, the displacement of the substrate along the conveying direction is divided by the area division step size and rounded down to obtain the area number. This area number is used as the area code for the printing area to be tested. In this way, the continuous displacement of the substrate along the conveying direction is divided into multiple sequentially arranged and non-overlapping spatial intervals. Each spatial interval corresponds to an area code, thus ensuring the uniqueness and stability of the area code.

[0028] In addition, when the substrate is changed or the equipment is restarted, the pulse count can be reset according to the physical zero mark on the substrate to reduce the impact of accumulated error on the accuracy of area coding.

[0029] S2. Before UV curing, obtain the dot area ratio of the printing area to be tested corresponding to the area code, and collect the first fluorescence response value.

[0030] The fluorescence response of UV inks before curing is typically related not only to the ink's own fluorescence properties but also to the ink coverage within the tested printing area. Dot area ratio reflects the proportion of ink coverage in that area to the total area; different dot area ratios result in different amounts of ink participating in fluorescence emission. Without obtaining the dot area ratio of the printed area corresponding to the area code, the fluorescence response values ​​between different areas may lack comparability due to differences in ink coverage, thus affecting the accuracy of subsequent curing degree assessment.

[0031] Therefore, before UV curing, this step obtains the prepress data of the corresponding printing area to be tested according to the area code, and extracts the dot area ratio of the printing area to be tested from it; at the same time, fluorescence excitation and fluorescence acquisition are performed on the printing area to be tested to obtain the first fluorescence response value in the uncured state, thereby providing benchmark data for subsequent calculation of the fluorescence response change value before and after curing.

[0032] In one specific embodiment, firstly, using the communication interface supported by the offset printing press control system, prepress plate-making data corresponding to the region code is obtained by using the region code as a query index. The prepress plate-making data includes rasterized image data of the printing area to be tested. Based on the rasterized image data, pixels covered by ink within the printing area to be tested are identified, and the ratio between the number of ink-covered pixels and the total number of pixels within the printing area to be tested is calculated. This ratio is used as the dot area ratio of the printing area to be tested.

[0033] Subsequently, fluorescence excitation spectroscopy tests were performed on the UV ink samples to be used. The UV ink samples were irradiated at different excitation wavelengths, and the corresponding fluorescence emission signals were collected under the same acquisition conditions to obtain the fluorescence emission intensity corresponding to each excitation wavelength. The excitation wavelength corresponding to the maximum fluorescence emission intensity was taken as the peak wavelength of the excitation light. The printing area to be tested was irradiated with excitation light whose peak wavelength was located within the fluorescence excitation spectrum of the ink, causing the UV ink in the printing area to be excited and produce fluorescence.

[0034] Finally, the fluorescence signal emitted by the ink is collected by a photodetector. This fluorescence signal is incident on the photosensitive surface of the photodetector, generating a photocurrent signal corresponding to the incident fluorescence intensity. This photocurrent signal is converted into a voltage signal by a transimpedance amplifier circuit, and then converted into a digital signal by an analog-to-digital converter circuit. This digital signal is used as the first fluorescence response value corresponding to the encoded region. This first fluorescence response value characterizes the fluorescence emission intensity of the printed area under specific excitation conditions before UV curing, serving as a benchmark for subsequent calculations of fluorescence response changes.

[0035] The photodetector can be calibrated using background signal and standard fluorescence reference sample before acquiring the first fluorescence response value, so as to reduce the influence of ambient light, background noise and detector response differences on the fluorescence acquisition results.

[0036] S3. After UV curing, according to the area code, match the same printing area to be tested, collect the second fluorescence response value of the same type as the first fluorescence response value, and obtain the ink layer thickness after curing and the absorption coefficient and scattering coefficient of the ink under the excitation light band and the emission light band.

[0037] Because the fluorescence response of UV ink changes after irradiation by a UV curing device, and the curing process may also cause changes in ink layer thickness, internal scattering structure, and optical properties, the degree of curing is judged solely based on the difference in fluorescence response before and after curing. Changes in ink layer thickness and optical properties may affect the propagation and attenuation of the fluorescence signal, making it difficult to accurately characterize the curing state of the ink based on the fluorescence response change value.

[0038] Therefore, this step involves spatially precise fluorescence acquisition of the same area after UV curing, and measurement of multi-dimensional optical parameters such as the ink layer thickness and absorption and scattering coefficients in the excitation and emission light bands. This separates the effects of changes in the physical structure and optical properties of the ink layer on the fluorescence signal, thus providing all the necessary optical input parameters for the subsequent construction of the optical path attenuation factor.

[0039] In one specific embodiment, firstly, after the UV curing device, when a test printing area consistent with the area code is identified according to the transport coding signal, the same excitation wavelength, irradiance and fluorescence collection solid angle as when the first fluorescence response value was collected are maintained, and a fluorescence signal is collected for the test printing area, and the collected fluorescence signal is used as a second fluorescence response value of the same type as the first fluorescence response value.

[0040] Subsequently, at the inspection station following the UV curing unit, the printed area to be tested, corresponding to the area code, is measured using a confocal color difference displacement sensor to obtain the thickness of the cured ink layer. Specifically, the confocal color difference displacement sensor projects a broadband focused beam onto the printed area to be tested through a dispersive objective lens. Since light of different wavelengths has different focusing positions along the optical axis of the sensor, when the upper surface of the ink layer and the surface of the substrate are located at their respective focusing positions, corresponding reflection spectral signals are formed. The two reflection spectral signals formed by reflection from the upper surface of the ink layer and the surface of the substrate are acquired separately, and the peak wavelengths of the two reflection spectra are extracted. and 2.

[0041] Based on the conversion relationship between peak wavelength and axial displacement, the axial displacement of the upper surface of the ink layer is calculated. Apparent axial displacement of the substrate surface Because the reflected signal from the surface of the substrate is received after propagating through the ink layer, the... and The difference between the two values ​​represents the apparent height difference affected by the ink's refractive index. Therefore, by measuring the actual physical thickness of a sample from the same batch as the ink being tested and treated with the same curing process, and using a confocal sensor to measure its apparent thickness, the ratio of the apparent thickness to the actual physical thickness is calculated as the equivalent refractive index neff. This apparent height difference is then corrected, and the cured ink layer thickness h is calculated using the following formula: .

[0042] The conversion relationship between peak wavelength and axial displacement is obtained through the following calibration method: A high-precision displacement stage is used to drive a standard diffuse reflection white board as a calibration target, moving it along the optical axis of the sensor with a fixed step distance. The peak wavelength of the reflection spectrum at each standard axial displacement position is recorded, thereby obtaining a set of displacement-wavelength calibration data. Based on this calibration data, the least squares method is used for linear fitting to establish a linear conversion relationship between peak wavelength and axial displacement. During measurement, the peak wavelength extracted in real time is substituted into this linear conversion relationship for calculation to obtain the corresponding axial displacement.

[0043] Finally, the polarized diffuse reflectance spectra of the printed area under test in the excitation and emission light bands are obtained, and the absorption and scattering coefficients of the ink are inverted by combining the thickness of the cured ink layer. Specifically, the printed area under test is illuminated with detection light in the excitation and emission light bands, respectively, and the intensity of parallel polarized reflected light is collected by a polarization spectrometer. Orthogonally polarized reflected light intensity The volume scattering component inside the ink layer was extracted from the reflected light signal using a polarization optics model. Its extraction expression is: .

[0044] in, and The polarization decomposition coefficients are determined based on a polarization optics model and are calibrated by measuring the polarization response of a standard diffuse reflector with known optical properties under a specific polarization path. Since the volume scattering within the ink layer exhibits depolarization characteristics while surface reflection maintains its polarization state, in a simplified implementation, we take... and That is, the orthogonal polarization component is directly used as the volume scattering component.

[0045] Subsequently, the thickness of the cured ink layer was measured. Volume scattering component As the inversion input, the absorption coefficients of the ink in the printing area under test are obtained by solving the radiative transfer equation in reverse. and scattering coefficient .

[0046] When solving the radiative transfer equation in reverse, the absorption coefficient is used and scattering coefficient As the variable to be inverted, a radiative transfer forward model based on the addition-multiplication method is established. This forward model is based on the variable to be inverted and the ink layer thickness. In addition to optical boundary parameters, the output yields the simulated volume scattering components. The deviation between the simulated volume scattering component and the measured volume scattering component is used as the optimization objective; an objective function is constructed. Its expression is: .

[0047] The variables to be inverted are iteratively adjusted using a nonlinear optimization algorithm until the objective function meets the convergence requirement, i.e. ,in, The convergence threshold is set to 10. -4 Up to 10 -6 The absorption and scattering coefficients at convergence are used as the inversion results. The optical boundary parameters include the reflection and transmission parameters at the ink-substrate interface.

[0048] It should be noted that during the iterative process of inversely solving the radiative transfer equation, a maximum number of iterations is set, such as 500. When the maximum number of iterations is reached and the convergence requirement is still not met, the iteration is forcibly terminated and the optimal solution of the current inverse equation is output.

[0049] S4. Calculate the difference between the first fluorescence response value and the second fluorescence response value as the fluorescence response change value; input the dot area ratio, the thickness of the ink layer after curing, and the absorption coefficient and scattering coefficient under the excitation light band and the emission light band into the photon transport probability model, and output the equivalent optical path; calculate the ratio of the equivalent optical path to the reference optical path as the optical path attenuation factor, and divide the fluorescence response change value by the optical path attenuation factor to obtain the target curing characteristic value.

[0050] Because changes in the ink layer thickness, absorption coefficient, and scattering coefficient after curing alter the propagation path length of fluorescent photons within the ink layer, the changes in fluorescence response before and after curing include not only the fluorescence changes caused by the curing reaction but also the effects of changes in ink layer structure and optical attenuation. Directly using the difference between the first and second fluorescence response values ​​as a measure of curing degree may lead to a lack of comparability between tested printed areas of different thicknesses or optical properties.

[0051] Therefore, this step simulates the transmission process of excitation photons and fluorescence photons in the printing area to be tested using a photon transport probability model, calculates the equivalent optical path corresponding to the current printing area to be tested, and constructs an optical path attenuation factor based on the ratio between the equivalent optical path and the reference optical path. Then, the optical path attenuation factor is used to compensate for the change in fluorescence response value to obtain the target curing characteristic value.

[0052] In one specific embodiment, firstly, a first fluorescence response value and a second fluorescence response value corresponding to the same region code are read, and the difference between the two is calculated, and the difference is used as the fluorescence response change value.

[0053] Subsequently, the dot area ratio, cured ink layer thickness, absorption coefficient and scattering coefficient in the excitation light band, and absorption coefficient and scattering coefficient in the emission light band corresponding to the encoded region are input into the photon transport probability model. This photon transport probability model is a numerical simulation model used to perform Monte Carlo photon tracing. It operates in the order of three-dimensional ink layer structure construction, virtual excitation photon tracing, virtual fluorescence photon tracing, and emission path length statistics, and outputs the equivalent optical path length based on the path length statistics of successfully emitted fluorescence photons.

[0054] Before running the model, a three-dimensional ink layer structure of the printing area to be tested is constructed. Specifically, the coverage area of ​​the excitation spot is divided into a two-dimensional planar discrete grid. The product of the total number of two-dimensional planar discrete grids and the dot area ratio is calculated, and the product is rounded down to obtain the number of grids covered by ink. The direction vector of the dot arrangement is determined according to the screen angle of the offset printing machine, and the spacing between the centers of adjacent dots is determined according to the screen ruling. A dot center coordinate sequence is generated on the two-dimensional planar discrete grid. Based on the center of each dot, the marking grid is expanded outward according to the dot shape, such as a circle, an ellipse, or the same as the screen ruling method of the offset printing machine, so that the number of marked grids is equal to or approximately equal to the number of grids covered by ink determined above, thereby forming a two-dimensional distribution pattern of ink dots.

[0055] After obtaining the two-dimensional distribution pattern of ink dots, the grid covered by ink is extended along a direction perpendicular to the surface of the substrate to the thickness of the cured ink layer, forming a three-dimensional ink dot column. The interior of the three-dimensional ink dot column is designated as the ink medium region, and the area below the un-ink-covered region is designated as the substrate medium region. An optical refractive index interface is established between the ink medium region, the substrate medium region, and the air. This optical refractive index interface is used to determine the reflection or transmission behavior of photons based on the refractive indices of the media on both sides of the interface when photons reach the medium boundary.

[0056] After completing the construction of the three-dimensional ink layer structure, virtual excitation photons are injected into the incident region corresponding to the excitation spot. The model performs Monte Carlo simulation to track the propagation distance, absorption events, and scattering events of the virtual excitation photons according to the absorption and scattering coefficients under the excitation light band. When the virtual excitation photon undergoes an absorption event in the ink medium region, the location of the absorption event is taken as the generation location of the virtual fluorescent photon, and a virtual fluorescent photon is generated at that location.

[0057] For the generated virtual fluorescent photons, the model continues to track their propagation process between the ink dot structure, the substrate interface, and the air interface according to the absorption and scattering coefficients in the emitted light band. When the virtual fluorescent photon reaches the optical refractive index interface during propagation, the model determines the reflection probability based on the photon's incident direction, incident angle, and the refractive indices of the media on both sides of the interface, and compares the reflection probability with a random number; if the random number is less than or equal to the reflection probability, it is determined that the virtual fluorescent photon has been reflected; otherwise, it is determined that the virtual fluorescent photon has been transmitted.

[0058] When a virtual fluorescent photon is emitted from the detection side and enters the fluorescence collection range, it is recorded as a successfully emitted fluorescent photon, and the cumulative path length traversed from the generation position to the emission position is recorded. After tracking no less than 10,000 virtual excitation photons, the model summarizes the cumulative path lengths of all successfully emitted fluorescent photons to form an effective optical path distribution, and the average path length of this effective optical path distribution is used as the equivalent optical path length of the current printing area to be tested.

[0059] The number of virtual photons injected into the model is determined through convergence testing. For example, the photon transport probability model is run with 10,000 virtual excitation photons as the initial number of virtual photons to obtain the corresponding equivalent optical path. Subsequently, 10,000 virtual excitation photons are added each time, and the model is run again, comparing the rate of change of the equivalent optical path obtained from two consecutive runs. When the rate of change of the equivalent optical path obtained from three consecutive runs is less than 1%, the equivalent optical path statistical result is considered to have reached a stable state, and the current number of virtual photons is used as the standard number of virtual photons for subsequent model runs.

[0060] After obtaining the equivalent optical path of the current printing area under test, the initial parameters of the current ink system in its uncured initial state are retrieved. These initial parameters include the initial ink layer thickness, initial absorption coefficient and initial scattering coefficient in the excitation light band, and initial absorption coefficient and initial scattering coefficient in the emission light band. These initial parameters are obtained through offline calibration of uncured ink standard samples. The initial parameters and the same dot area ratio as the current printing area under test are input into the same photon transport probability model. The model is run using the same virtual photon count, optical boundary conditions, and fluorescence collection conditions as the current printing area under test. The average path length of successfully emitted fluorescent photons under these conditions (i.e., when the ink is in its uncured initial state) is calculated and used as the reference optical path length.

[0061] Finally, the ratio between the equivalent optical path and the reference optical path of the current printing area to be tested is calculated, and this ratio is used as the optical path attenuation factor. If the calculated optical path attenuation factor is less than the minimum threshold, such as 0.001, the optical path attenuation factor is set to the minimum threshold to avoid division by zero anomalies.

[0062] like Figure 2 As shown, the horizontal axis represents the dot area ratio, the vertical axis represents the ink layer thickness after curing, and the thermochromic gradation and contour lines represent the optical path attenuation factor of the current test area relative to the reference state. With the increase of dot area ratio and ink layer thickness, the optical path attenuation factor generally increases, indicating that the ink coverage state and film structure jointly affect the effective propagation path of fluorescent photons. Therefore, in the process of extracting target curing features, it is necessary to use the optical path attenuation factor to compensate for changes in fluorescence response, in order to reduce the interference of non-curing factors on the judgment of the degree of curing.

[0063] Therefore, the target curing characteristic value is obtained by dividing the fluorescence response change value by the optical path attenuation factor. This target curing characteristic value represents the degree of fluorescence response change after optical path attenuation compensation and is used as input features for subsequent curing degree identification models.

[0064] S5. Input the target solidification feature value into the solidification degree identification model, and output the solidification rate prediction value corresponding to the region code.

[0065] Considering that the target curing feature value can characterize the degree of fluorescence response change in the printed area after optical path attenuation compensation, and that the degree of fluorescence response change is related to the curing state of the UV ink, the correspondence between the target curing feature value and the actual curing rate may differ under different ink formulations and substrate systems. Therefore, this step uses a pre-trained curing degree recognition model to perform regression processing on the target curing feature value and outputs the predicted curing rate value corresponding to the region code.

[0066] In one specific embodiment, the solidification degree identification model adopts a univariate quadratic polynomial regression model, the model function of which is expressed as: .

[0067] in, For the target solidified feature value, This is the predicted curing rate. , , These are the parameters of the model to be trained.

[0068] The training process for the model parameters is as follows: Standard ink samples with different curing states are prepared, and the target curing characteristic values ​​and the actual curing rate measured by differential scanning calorimetry are obtained for each sample to form training samples. The mean square error loss function is constructed, and the minimum value of the loss function is solved using the least squares method to calculate the parameters that minimize the loss function. , , Find the closed-form solution and complete the model training.

[0069] During online detection, the obtained target solidification feature value is input into the trained solidification degree recognition model. The model outputs a predicted solidification rate value corresponding to the target solidification feature value, and stores the predicted solidification rate value in association with the corresponding region code.

[0070] like Figure 3 As shown, the scatter plots represent the actual curing rate samples obtained by differential scanning calorimetry, the orange curve represents the quadratic regression prediction result output by the curing degree identification model, the light blue area represents the 95% residual band, and the horizontal dashed line represents the preset curing qualification threshold of 90%. As the target curing feature value increases, the predicted curing rate value rises overall and gradually approaches or exceeds the qualification threshold, indicating that the target curing feature value after optical path attenuation compensation can effectively map the predicted curing rate value corresponding to the region code.

[0071] Furthermore, after the predicted curing rate is output, it is compared with a preset curing pass threshold. When the predicted curing rate is lower than the curing pass threshold, it is determined that the printed area to be tested corresponding to the area code has insufficient curing, and a curing abnormality alarm signal is output; when the predicted curing rate is not lower than the curing pass threshold, it is determined that the printed area to be tested corresponding to the area code meets the curing requirements.

[0072] The method for obtaining the curing qualification threshold is as follows: the adhesion level of samples with different curing rates is determined by cross-cut adhesion test, and the curing rate corresponding to the adhesion level meeting the minimum process requirements of the product, such as 4B level, is taken as the curing qualification threshold, such as 90%, and the test is recalibrated when the ink system or substrate material is changed.

[0073] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0074] Those skilled in the art will recognize that the algorithmic steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0075] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0076] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0077] Finally, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for intelligent detection of the curing degree of UV ink in an offset printing machine, characterized in that, include: The area code of the printing area to be tested is determined based on the pulse count of the encoder and the size of the excitation spot on the offset printing machine. Before UV curing, the dot area ratio of the printing area to be tested corresponding to the area code is obtained, and the first fluorescence response value is collected. After UV curing, the same printing area to be tested is matched according to the regional code, and a second fluorescence response value of the same type as the first fluorescence response value is collected. The thickness of the ink layer after curing and the absorption coefficient and scattering coefficient of the ink under the excitation light band and the emission light band are obtained. The difference between the first fluorescence response value and the second fluorescence response value is calculated as the fluorescence response change value; the dot area ratio, the thickness of the ink layer after curing, and the absorption coefficient and scattering coefficient under the excitation light band and the emission light band are input into the photon transport probability model to output the equivalent optical path; the ratio of the equivalent optical path to the reference optical path is calculated as the optical path attenuation factor, and the fluorescence response change value is divided by the optical path attenuation factor to obtain the target curing characteristic value. The photon transport probability model is a numerical simulation model used to perform Monte Carlo photon tracing. It runs in the order of three-dimensional ink layer structure construction, virtual excitation photon tracing, virtual fluorescence photon tracing, and emission path length statistics, and outputs the equivalent optical path based on the path length statistics of successfully emitted fluorescence photons. The method for obtaining the reference optical path is as follows: Obtain the initial ink layer thickness of the current ink system in the uncured initial state, as well as the initial absorption coefficient and initial scattering coefficient in the excitation light band and emission light band in the uncured initial state; The initial ink layer thickness, the same dot area ratio as the current printing area to be tested, and the initial absorption coefficient and initial scattering coefficient are input into the photon transport probability model. Under the same simulated photon number and optical boundary conditions, the equivalent optical path under these conditions is output as the reference optical path. Input the target solidification feature value into the solidification degree identification model, and output the predicted solidification rate value corresponding to the region code; The method for obtaining the predicted curing rate is as follows: The target solidification feature value is input into the solidification degree recognition model, which is obtained by regression training from multiple sets of target solidification feature value samples with known solidification rates. Obtain the predicted curing rate value corresponding to the region code output by the curing degree identification model.

2. The intelligent detection method for the curing degree of UV ink in an offset printing machine as described in claim 1, characterized in that, The method for determining the region code of the printing area to be tested is as follows: The pulse signals output by the encoder of the offset printing machine are counted, and the pulse count is multiplied by the encoder resolution to obtain the displacement of the substrate along the conveying direction. The size of the excitation spot in the transport direction when collecting fluorescence signals is set as the region division step size; Divide the displacement by the region division step size and round down to obtain the region number, which is then used as the region code for the printing area to be tested.

3. The intelligent detection method for the curing degree of UV ink in an offset printing machine as described in claim 1, characterized in that, The method for obtaining the dot area ratio and the first fluorescence response value is as follows: The prepress plate-making data corresponding to the code of this area is obtained from the offset printing machine control system through the communication interface, and the dot area ratio is extracted from it. The printed area under test is irradiated with excitation light whose peak wavelength is within the excitation spectrum of ink fluorescence. The emitted fluorescence is collected and photoelectric conversion is performed to obtain the first fluorescence response value.

4. The intelligent detection method for the curing degree of UV ink in an offset printing machine as described in claim 1, characterized in that, The method for obtaining the second fluorescence response value is as follows: After the UV curing device, when the test printing area that matches the area code is identified according to the transport coding signal, the same excitation wavelength, irradiance and fluorescence collection solid angle as when the first fluorescence response value was collected are maintained, and the fluorescence signal is collected as the second fluorescence response value.

5. The intelligent detection method for the curing degree of UV ink in an offset printing machine as described in claim 1, characterized in that, The acquisition of the cured ink layer thickness and the absorption and scattering coefficients of the ink under the excitation and emission light bands specifically includes: At the inspection station after the UV curing unit, the printing area to be tested corresponding to the area code is measured using a confocal color difference displacement sensor to obtain the thickness of the ink layer after curing. The polarization diffuse reflectance spectrum of the printing area under test is obtained in the excitation light band and the emission light band. Based on the polarization diffuse reflectance spectrum, the volume scattering component inside the ink layer is extracted using a polarization optics model. Combined with the thickness of the cured ink layer and the volume scattering component, the absorption coefficient and scattering coefficient of the ink in the printing area under test in the excitation light band and the emission light band are obtained by inversely solving the radiative transfer equation.

6. The intelligent detection method for the curing degree of UV ink in an offset printing machine as described in claim 1, characterized in that, The method for obtaining the equivalent optical path is as follows: The dot area ratio, the thickness of the cured ink layer, the absorption coefficient and scattering coefficient under the excitation light band, and the absorption coefficient and scattering coefficient under the emission light band are input into the photon transport probability model. Run the photon transport probability model, construct the spatial distribution structure of ink dots based on the dot area ratio inside the model, set the spatial longitudinal boundary of the model based on the thickness of the ink layer after curing, inject virtual excitation photons into the spatial distribution structure, and track the excitation photons to the absorption position based on the absorption coefficient and scattering coefficient of the excitation light band to generate virtual fluorescent photons. Based on the absorption and scattering coefficients of the emitted light band, the path lengths of virtual fluorescent photons that are transmitted and scattered between the ink dot structure, ink layer and substrate interface are tracked. The path length data of all emitted fluorescent photons are statistically analyzed to obtain the effective optical path distribution. The expected value of the effective optical path distribution is taken as the equivalent optical path.

7. The intelligent detection method for the curing degree of UV ink in an offset printing machine as described in claim 6, characterized in that, The construction of the spatial distribution structure of ink dots based on the dot area ratio specifically includes: Establish a two-dimensional planar discrete grid corresponding to the irradiation range of the excitation spot; Based on the dot area ratio and the screen parameters of the offset printing machine, a two-dimensional distribution pattern of ink dots is generated on a two-dimensional discrete grid; wherein, the proportion of the number of grids covered by ink to the total number of grids is equal to the dot area ratio; By extending the cured ink layer thickness in a direction perpendicular to the surface of the substrate, a three-dimensional ink dot column is constructed from the grid covered by ink in a two-dimensional distribution pattern. The interior of the three-dimensional ink dot column is set as the ink medium region, the area below the grid not covered by ink is set as the substrate medium region, and an optical refractive index interface is set at the junction of the ink medium region and the substrate medium region, thereby forming the spatial distribution structure of the ink dots.

8. The intelligent detection method for the curing degree of UV ink in an offset printing machine as described in claim 1, characterized in that, After outputting the predicted curing rate value corresponding to the region code, the following is also included: The predicted curing rate is compared with the curing qualification threshold; if the predicted curing rate is lower than the curing qualification threshold, an abnormal curing alarm signal is output.

Images