CN-122008689-A - Gravure printing process parameter optimization method and system based on tension fluctuation analysis

CN122008689ACN 122008689 ACN122008689 ACN 122008689ACN-122008689-A

Abstract

The application provides a gravure printing process parameter optimization method and a gravure printing process parameter optimization system based on tension fluctuation analysis, which belong to the technical field of printing process parameter optimization, and the method comprises the steps of collecting tension fluctuation signals in the operation process of a gravure printing machine and collecting corresponding printing process parameters; the method comprises the steps of preprocessing tension fluctuation signals and extracting features to obtain multi-dimensional feature vectors, inputting printing process parameters into a preset parameter-tension prediction model to obtain predicted multi-dimensional feature vectors, calculating residual errors of the multi-dimensional feature vectors and the predicted multi-dimensional feature vectors, identifying abnormal process parameters through residual error analysis, adjusting and optimizing the abnormal process parameters by using an optimization algorithm based on physical constraint of the printing process parameters by taking minimized tension fluctuation and overprinting errors as optimization targets, generating an optimized process parameter set, and updating the optimized process parameter set to a gravure printing process machine to optimize the printing process parameters.

Inventors

ZHAO SIYUAN
Zheng Shuke
WANG DONGMIN
LI JUNTAO
WANG LIQIN

Assignees

河北显丰包装材料有限公司

Dates

Publication Date: 20260512
Application Date: 20260127

Claims (10)

1. The intaglio printing process parameter optimization method based on tension fluctuation analysis is characterized by comprising the following steps of: collecting tension fluctuation signals of the intaglio printing press in the running process, and synchronously collecting corresponding printing process parameters; preprocessing and extracting features of the tension fluctuation signal to obtain a multi-dimensional feature vector, wherein the multi-dimensional feature vector comprises time domain features and frequency domain features; Inputting the printing process parameters into a preset parameter-tension prediction model to obtain a predicted multidimensional feature vector; Calculating residual errors of the multi-dimensional feature vector and the predicted multi-dimensional feature vector, and identifying abnormal process parameters through residual error analysis; Taking minimized tension fluctuation and overprinting error as optimization targets, adjusting and optimizing the abnormal process parameters by using an optimization algorithm based on physical constraint of the printing process parameters, and generating an optimized process parameter set; updating the optimized process parameter set to the intaglio printing press to optimize the printing process parameters.
2. The method for optimizing gravure printing process parameters based on tension fluctuation analysis according to claim 1, wherein the preprocessing and feature extraction of the tension fluctuation signal to obtain a multidimensional feature vector comprises the following steps: performing outlier rejection and moving average filtering processing on the tension fluctuation signal to obtain a steady-state tension signal; calculating statistical characteristics of the steady-state tension signals as time domain characteristics based on the steady-state tension signals, wherein the statistical characteristics comprise standard deviation, peak-to-peak value, root mean square value and waveform factor of the signals; performing fast Fourier transform on the steady-state tension signal to obtain a frequency spectrum; extracting the frequency corresponding to the energy duty ratio of the preset characteristic frequency band and the spectral peak with the amplitude larger than the first frequency threshold from the frequency spectrum as the frequency domain characteristic; And combining the time domain features and the frequency domain features based on a predetermined sequence to form the multi-dimensional feature vector.
3. A method for optimizing intaglio printing process parameters based on tension fluctuation analysis according to claim 2, wherein said combining extracted time domain features with frequency domain features according to a predetermined order comprises: Respectively carrying out normalization processing on the time domain features and the frequency domain features; Arranging the normalized time domain features according to a fixed sequence of standard deviation, peak-to-peak value, root mean square value and waveform factor, arranging the normalized frequency domain features according to a sequence of the energy duty ratio of the characteristic frequency band and the significant spectrum peak frequency from small to large, and splicing to form a basic feature vector; Calculating the correlation coefficient of the variance and tension fluctuation of each component in the basic feature vector; And removing redundant characteristic components with the variance smaller than a preset first threshold and the correlation coefficient smaller than a preset second threshold to obtain the multidimensional characteristic vector.
4. The method for optimizing gravure printing process parameters based on tension fluctuation analysis according to claim 1, wherein the calculating the residual error between the multi-dimensional feature vector and the predicted multi-dimensional feature vector, and identifying the abnormal process parameters by residual error analysis, comprises: Arranging the multi-dimensional feature vector and the predicted multi-dimensional feature vector according to the same sequence, subtracting the elements by elements and taking the absolute value to obtain an initial residual vector; comparing each residual value in the initial residual vector with a preset abnormality judgment threshold value of a corresponding characteristic component, and marking the characteristic component with the residual value larger than the corresponding preset abnormality judgment threshold value as an abnormality characteristic; and outputting one or more process parameters directly related to the abnormal characteristics as the abnormal process parameters according to a preset abnormal characteristic-process parameter lookup table.
5. The method according to claim 4, wherein before comparing each residual value in the initial residual vector with a preset anomaly determination threshold value of a corresponding feature component, further comprising: adjusting the preset abnormal judgment threshold value of each characteristic component according to the accumulated running time and the abrasion state of the preset component, wherein the preset component comprises a plate roller and a scraper; Acquiring the cumulative revolution of the plate roller, the pressure and angle historical data of the scraper, and calculating to obtain a comprehensive health index through a preset abrasion model; and if the comprehensive health index is smaller than the preset health threshold, the preset abnormality judgment threshold of the corresponding characteristic component is improved based on the first step length.
6. The method for optimizing gravure printing process parameters based on tension fluctuation analysis according to claim 1, wherein the optimizing the abnormal process parameters by using an optimization algorithm based on physical constraints of the printing process parameters with the aim of minimizing tension fluctuation and overprinting errors to generate an optimized process parameter set comprises: Constructing a comprehensive objective function, wherein the comprehensive objective function aims at minimizing tension fluctuation and overprinting errors; Determining an adjustment range of each abnormal process parameter to obtain the physical constraint, wherein the adjustment range comprises upper and lower limits of the abnormal process parameter and a linear or nonlinear coupling relation between the parameters; In the search space defined by the physical constraint, carrying out iterative search on the abnormal process parameters by using a genetic algorithm and taking the minimized comprehensive objective function as a target; and when the genetic algorithm reaches a convergence condition, outputting the currently searched optimal parameter combination as the optimized process parameter set.
7. The method for optimizing gravure printing process parameters based on tension fluctuation analysis according to claim 1, wherein the calculating the residual error between the multi-dimensional feature vector and the predicted multi-dimensional feature vector, after identifying the abnormal process parameters by residual error analysis, further comprises: Calculating the integral residual norms of the multi-dimensional feature vector and the predicted multi-dimensional feature vector; Outputting a model parameter optimization requirement for the parameter-tension prediction model in response to the integral residual norm being greater than a preset model update threshold; And responding to the model parameter optimization requirement, and carrying out parameter optimization on the parameter-tension prediction model by using a particle swarm optimization algorithm, wherein the parameter-tension prediction model is a neural network model.
8. The method for optimizing parameters of a gravure printing process based on tension fluctuation analysis according to claim 7, wherein before the parameter-tension prediction model is optimized by using a particle swarm optimization algorithm, the method further comprises: Determining the search space of the particle swarm optimization algorithm according to the number and the value range of parameters to be optimized in the neural network model; And determining the super-parameter combination of the particle swarm optimization algorithm according to the magnitude of the integral residual norm.
9. The method for optimizing gravure printing process parameters based on tension fluctuation analysis according to claim 8, wherein the super-parameters comprise inertial weights and learning factors, wherein the determining the super-parameters of the particle swarm optimization algorithm according to the magnitude of the integral residual norms comprises: If the integral residual norm is larger than or equal to a preset first residual threshold, reducing the reference value of the inertia weight based on a first adjustment step length, and improving the reference value of the learning factor based on a second adjustment step length; if the integral residual norm is smaller than a preset second residual threshold, the reference value of the inertia weight is increased based on the third adjustment step length, and the reference value of the learning factor is reduced based on the fourth adjustment step length.
10. An intaglio printing process parameter optimization system based on tension fluctuation analysis, comprising: The data acquisition module is used for acquiring tension fluctuation signals of the intaglio printing press in the running process and synchronously acquiring corresponding printing process parameters; The feature extraction module is used for preprocessing and extracting features of the tension fluctuation signal to obtain a multi-dimensional feature vector, wherein the multi-dimensional feature vector comprises time domain features and frequency domain features; the model prediction module is used for inputting the printing process parameters into a preset parameter-tension prediction model to obtain a predicted multidimensional feature vector; The abnormal parameter module is used for calculating residual errors of the multi-dimensional characteristic vector and the predicted multi-dimensional characteristic vector, and identifying abnormal technological parameters through residual error analysis; The adjustment optimization module is used for adjusting and optimizing the abnormal process parameters by using an optimization algorithm based on physical constraint of the printing process parameters by taking the minimized tension fluctuation and overprinting error as optimization targets to generate an optimized process parameter set; And the parameter updating module is used for updating the optimized process parameter set to the intaglio printing press to optimize the printing process parameter.

Description

Gravure printing process parameter optimization method and system based on tension fluctuation analysis Technical Field The application relates to the technical field of printing process parameter optimization, in particular to a gravure printing process parameter optimization method and system based on tension fluctuation analysis. Background In the existing intaglio printing process parameter adjustment process, a mode of manual experience judgment and trial and error is adopted, and an operator observes appearance quality of a printed matter, such as color deviation and overprinting precision, through own working experience to judge whether the printing process parameter is suitable. If the printed matter is found to have problems, parameters such as unreeling tension, reeling tension, printing unit speed, unit-to-unit tension, drying temperature and the like are gradually adjusted, and relatively suitable parameter combinations are found through multiple tests. This approach is inefficient and requires extensive experimentation and adjustment. In addition, a simple sensor is adopted to monitor partial parameters, such as tension force, and the monitoring mode can only acquire single parameter values, lacks comprehensive analysis on correlations among parameters and tension fluctuation conditions, cannot accurately identify abnormal process parameters, and cannot effectively optimize printing process parameters. Therefore, there is a need for a method and system for optimizing gravure process parameters based on tension fluctuation analysis. Disclosure of Invention In order to solve the technical problems, the application provides a gravure printing process parameter optimization method and system based on tension fluctuation analysis. In a first aspect of the embodiment of the present application, there is provided a method for optimizing parameters of an intaglio printing process based on tension fluctuation analysis, including: collecting tension fluctuation signals of the intaglio printing press in the running process, and synchronously collecting corresponding printing process parameters; preprocessing and extracting features of the tension fluctuation signal to obtain a multi-dimensional feature vector, wherein the multi-dimensional feature vector comprises time domain features and frequency domain features; Inputting the printing process parameters into a preset parameter-tension prediction model to obtain a predicted multidimensional feature vector; Calculating residual errors of the multi-dimensional feature vector and the predicted multi-dimensional feature vector, and identifying abnormal process parameters through residual error analysis; Taking minimized tension fluctuation and overprinting error as optimization targets, adjusting and optimizing the abnormal process parameters by using an optimization algorithm based on physical constraint of the printing process parameters, and generating an optimized process parameter set; updating the optimized process parameter set to the intaglio printing press to optimize the printing process parameters. In a second aspect of the embodiment of the present application, there is provided an intaglio printing process parameter optimization system based on tension fluctuation analysis, including: The data acquisition module is used for acquiring tension fluctuation signals of the intaglio printing press in the running process and synchronously acquiring corresponding printing process parameters; The feature extraction module is used for preprocessing and extracting features of the tension fluctuation signal to obtain a multi-dimensional feature vector, wherein the multi-dimensional feature vector comprises time domain features and frequency domain features; the model prediction module is used for inputting the printing process parameters into a preset parameter-tension prediction model to obtain a predicted multidimensional feature vector; The abnormal parameter module is used for calculating residual errors of the multi-dimensional characteristic vector and the predicted multi-dimensional characteristic vector, and identifying abnormal technological parameters through residual error analysis; The adjustment optimization module is used for adjusting and optimizing the abnormal process parameters by using an optimization algorithm based on physical constraint of the printing process parameters by taking the minimized tension fluctuation and overprinting error as optimization targets to generate an optimized process parameter set; And the parameter updating module is used for updating the optimized process parameter set to the intaglio printing press to optimize the printing process parameter. In a third aspect of the embodiment of the present application, there is provided an electronic device including a memory, a processor, and a computer program stored in the memory and running on the processor, the processor implementing the steps of the above-described