A special paper texture imprinting robot array dynamic adjustment system

Abstract

The application discloses a special paper texture embossing robot array dynamic adjustment system, and relates to the technical field of special paper texture embossing. The special paper texture embossing robot array dynamic adjustment system can realize real-time acquisition of paper three-dimensional microscopic topography parameters, prediction of the optimal embossing depth and pressure based on a texture embossing plasticity index, differential embossing of different regions, guarantee of texture forming quality, real-time compensation and global equalization of the embossing force through force sensing monitoring, stress field calculation and distributed cooperative control algorithm.

Description

Technical Field

[0001] This invention relates to the field of special paper texture embossing, and more specifically, to a dynamic adjustment system for a special paper texture embossing robot array. Background Technology

[0002] Texture embossing for specialty papers is an important process for enhancing the added value and visual appeal of paper, and is widely used in high-end packaging, art paper, and functional paper products. Current technologies typically rely on mechanical embossing equipment or automated embossing systems with fixed parameters. While robot arrays are sometimes introduced for collaborative operation, the overall process still relies heavily on preset embossing parameters, lacking the ability to precisely perceive and dynamically adjust for differences in the paper's microstructure. Furthermore, traditional systems lack effective means of acquiring key morphological parameters such as fiber distribution, pore structure, and initial roughness on the paper surface, making it difficult to achieve precise regional control of embossing depth and pressure. This results in inconsistent texture formation across different areas. In addition, during multi-robot collaborative embossing, current technologies generally lack real-time stress field monitoring and balancing mechanisms, easily leading to localized embossing force imbalances, which can cause paper deformation, texture distortion, and even material damage.

[0003] On the other hand, specialty paper is significantly affected by ambient temperature and humidity, as well as its own moisture content, during the printing process, easily leading to expansion or contraction caused by moisture-heat coupling. Existing printing systems typically do not consider the dynamic changes in the paper's moisture and heat state and lack path scheduling and pressure correction strategies based on moisture and heat characteristics, making it difficult to effectively avoid texture misalignment and accuracy deviations. Furthermore, most existing technologies employ open-loop control, lacking high-precision detection and feedback optimization mechanisms for printing results. This prevents iterative correction of printing parameters based on the actual texture formation effect, hindering further improvement in overall printing accuracy and consistency. Therefore, there is an urgent need for an intelligent printing system that integrates multi-scale sensing, stress equalization control, moisture-heat adaptive scheduling, and closed-loop optimization capabilities to meet the demands of high-quality specialty paper texture processing. Summary of the Invention

[0004] The purpose of this invention is to provide a dynamic adjustment system for a special paper texture imprinting robot array. This system solves the problem that traditional systems lack effective means to obtain key morphological parameters such as fiber distribution, pore structure, and initial roughness of the paper surface, making it difficult to achieve precise regional control of imprinting depth and pressure. This results in inconsistent imprinting textures in different areas, and the lack of a real-time stress field monitoring and balancing control mechanism easily leads to local imprinting force imbalance, which in turn causes paper deformation, texture distortion, and even material damage, failing to meet usage requirements.

[0005] This invention achieves the above objective through the following technical solution: a dynamic adjustment system for a special paper texture imprinting robot array, the system comprising: Multi-scale morphology perception and imprinting control unit, robot array stress equalization control unit, paper wet and heat coupling path scheduling unit, and special paper texture full-dimensional adjustment unit. The multi-scale topography perception imprint control unit constructs a texture imprint depth prediction and adjustment mechanism based on multi-scale surface topography perception, which is used to dynamically control the imprint depth and output pressure of the robot array. The robot array stress equalization control unit constructs a dynamic equalization control method for imprinting force based on the collaborative allocation of robot array stress field, thereby realizing real-time collaborative compensation and global equalization regulation of array imprinting force. The paper moisture and heat coupling path scheduling unit constructs an adaptive scheduling method for printing paths based on the prediction of paper moisture and heat coupling state, and dynamically adjusts the printing operation sequence and path planning strategy of the robot array. The special paper texture full-dimensional adjustment unit, in coordination with the operation mechanism and control data of the above three units, realizes full-dimensional dynamic adjustment of the special paper texture imprinting robot array, completes differentiated and consistent texture imprinting operations of special paper, and adapts to the imprinting process requirements and target texture specifications of different types of special paper.

[0006] Furthermore, the multi-scale topography sensing imprint control unit specifically includes the following steps: Equipped with a non-contact multi-scale structured light scanning module, the multi-scale structured light scanning module is deployed at the front end of the imprinting robot array to perform micron-level three-dimensional morphology full-domain scanning on the surface of special paper, covering the entire area of ​​the paper imprinting operation; Accurately acquiring three-dimensional morphological parameters such as fiber distribution density, micropore size distribution, and initial texture roughness at various locations on the paper surface provides comprehensive and high-precision basic data support for the dynamic control of printing depth and pressure.

[0007] Furthermore, the multi-scale topography sensing imprint control unit also includes the following steps: Built-in morphology spectrum feature extraction algorithm and imprint depth prediction model; The morphology spectrum feature extraction algorithm performs normalization preprocessing and multi-dimensional feature fusion on the three-dimensional morphology parameters. It assigns corresponding weight coefficients to each parameter in combination with the characteristics of special paper material and the requirements of printing process, and constructs a texture printing plasticity index that characterizes the printing deformation ability of different areas of paper. The imprint depth prediction model uses the texture imprint plasticity index as the core input feature, and combines the material type of the special paper and the process parameters of the target imprint texture fineness level to calculate the optimal imprint depth and imprint pressure combination parameters at each position of the paper in real time.

[0008] Furthermore, the multi-scale topography sensing imprint control unit includes the following steps: Establish real-time communication connections with the motion control system and pressure regulation system of the robot array; Using the calculated optimal combination of printing depth and printing pressure as control commands, each printing robot dynamically adjusts the travel distance and output pressure of the printing head in different areas of the paper, achieving precise control of the regionally differentiated printing depth of different positions on the paper, matching the actual printing deformation capacity of each area of ​​the paper, and ensuring the forming quality of the texture printing.

[0009] Furthermore, the robot array stress equalization control unit includes the following steps: A force sensing unit is configured between the end effector and the impression head of each imprinting robot; The force sensing unit communicates with the robot control system in real time. During the entire printing process, it collects the contact force and micro-vibration acceleration information of each robot printing head and the paper at a preset sampling frequency, providing real-time detection data for the balanced control of printing force.

[0010] Furthermore, the robot array stress equalization control unit also includes the following steps: Built-in filtering and noise reduction algorithm and stress field distribution model of the imprinting area; The filtering and denoising algorithm filters the collected raw information of contact force and micro-vibration acceleration, eliminating invalid data caused by interference factors such as environmental vibration and equipment operation, and obtaining effective contact force data that can truly reflect the imprinting contact state. The stress field distribution model of the imprinting area uses effective contact force as the basic data source. It combines the real-time detected paper tension state and the imprinting position coordinates of each robot imprinting head to calculate the continuous overall imprinting stress distribution in the entire imprinting area of ​​the paper through global interpolation, accurately representing the real-time stress state of each position of the paper during the imprinting process.

[0011] Furthermore, the robot array stress equalization control unit presets an imprinting stress threshold based on the material strength of the special paper and the imprinting process requirements. It then compares the real-time calculated imprinting stress distribution with the imprinting stress threshold position by position to accurately determine abnormal areas where the imprinting pressure is too high or too low. The unit also incorporates a distributed collaborative control algorithm. Using the stress field distribution in the printing area as the algorithm input, it identifies areas of abnormal pressure and adjacent printing robots. It controls adjacent robots to perform real-time force compensation adjustments based on the stress field distribution, calculates the corresponding pressure adjustment amount, and updates the output printing pressure of each robot in real time, forming a dynamic pressure balancing mechanism for the robot array to achieve uniform distribution of stress across the entire paper during the printing process.

[0012] Furthermore, the paper moisture-heat coupling path scheduling unit includes the following steps: A temperature and humidity detection module is installed in the pre-printing section of the special paper conveying channel; The temperature and humidity detection module includes a humidity sensor array and an infrared temperature detection module, and the detection area covers the entire width of the paper conveying process. The paper surface is scanned and detected position by position according to a preset fixed sampling interval, and the moisture content and surface temperature parameters of different areas of the paper are obtained in real time, so as to realize the full-area and real-time detection of the paper's moisture and heat status.

[0013] Furthermore, the paper moisture-heat coupling path scheduling unit incorporates a moisture-heat coupling deformation prediction model and a path planning algorithm. The moisture-heat coupling deformation prediction model takes the paper's moisture-heat state parameters as input, combines the fiber material characteristics of special paper, predicts the local linear expansion rate of paper affected by moisture and heat during the printing process, and classifies the paper printing area into three types based on the preset linear expansion rate threshold: high humidity expansion, drying shrinkage, and normal. The path planning algorithm dynamically adjusts the printing sequence and path planning strategy of the robot array based on the type of printing area, the arrangement of the robot array, and the work efficiency. At the same time, it matches the corresponding printing pressure correction value for different areas to achieve precise matching between printing pressure and paper humidity and heat conditions, thereby avoiding texture deformation and misalignment problems caused by changes in paper temperature and humidity at the process path level.

[0014] Furthermore, a closed-loop feedback correction mechanism is also included: The closed-loop feedback correction link is equipped with a high-precision texture detection device. After the printing operation is completed, the texture of the printed paper surface is detected over the entire area to obtain the parameters of the actual texture depth, clarity and continuity. The actual detection results are compared with the target texture parameters position by position to calculate the texture deviation value. The closed-loop feedback correction process also incorporates an iterative optimization algorithm. Based on the magnitude and distribution of the texture deviation value, combined with the previous imprinting parameter adjustment data, it updates the adjustment coefficients and threshold parameters of the imprinting depth, imprinting pressure, and imprinting path. The updated parameters are then fed back to the control systems of the robot array in real time, realizing closed-loop dynamic adjustment of the imprinting parameters, continuously optimizing the imprinting parameter configuration, and improving the forming accuracy and consistency of the imprinted texture.

[0015] The beneficial effects of this invention are as follows: 1. Through the multi-scale morphology perception and imprint control unit, the three-dimensional micro-morphology parameters of the paper are acquired in real time, and the optimal imprinting depth and pressure are predicted based on the texture imprinting plasticity index, so as to realize differentiated imprinting in each region and ensure the quality of texture forming.

[0016] 2. The robot array stress equalization control unit achieves real-time compensation and global balance of imprinting force through force sensing monitoring, stress field calculation and distributed collaborative control algorithm, so as to avoid excessive or insufficient local imprinting force and improve imprinting consistency.

[0017] 3. The paper humidity and heat coupling path scheduling unit can detect the paper temperature and humidity status in real time, predict the local linear expansion rate, and dynamically adjust the imprinting path and pressure correction value to avoid texture deformation, misalignment and uneven local imprinting.

[0018] 4. The closed-loop feedback correction process can perform full-area detection of the paper texture after printing and update the printing parameters in combination with iterative optimization algorithms to achieve continuous optimization of the printing process and significantly improve the consistency of texture depth, clarity and continuity.

[0019] 5. The system can automatically adjust the printing parameters and path according to the material characteristics and process requirements of different types of specialty paper, taking into account both differentiation and consistency, and meeting the diverse printing process needs. Attached Figure Description

[0020] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments of this application and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a flowchart of the multi-scale morphology sensing imprint control unit of the present invention; Figure 2 This is a flowchart of the robot array stress equalization control unit of the present invention; Figure 3 This is a flowchart of the paper moisture and heat coupling path scheduling unit of the present invention. Detailed Implementation

[0021] The present application will now be described in further detail with reference to the accompanying drawings. It should be noted that the following specific embodiments are only used to further illustrate the present application and should not be construed as limiting the scope of protection of the present application. Those skilled in the art can make some non-essential improvements and adjustments to the present application based on the above application content. Example

[0022] Please see Figure 1-3 This invention provides a technical solution: a dynamic adjustment system for a special paper texture imprinting robot array, the system comprising: Multi-scale morphology perception and imprinting control unit, robot array stress equalization control unit, paper moisture and heat coupling path scheduling unit, and special paper texture full-dimensional adjustment unit. A multi-scale topography-sensing imprint control unit is used to construct a texture imprint depth prediction and adjustment mechanism based on multi-scale surface topography perception, and to dynamically control the imprint depth and pressure of the robot array. Among them, multi-scale surface morphology perception detects and identifies the surface morphological features of special paper at different scales (such as macroscopic and microscopic), and obtains detailed information such as texture and undulation of the paper surface at different scales; texture imprinting depth prediction and adjustment mechanism establishes a mechanism that can predict the texture imprinting depth based on the information obtained from multi-scale surface morphology perception, and dynamically adjust it according to the prediction results, thereby dynamically controlling the imprinting depth and pressure of the robot array. The robot array stress equalization control unit is used to construct a dynamic equalization control method for imprinting force based on the collaborative distribution of robot array stress field, so as to realize the collaborative compensation and equalization of array imprinting force. Among them, the robot array stress field collaborative allocation takes into account the interaction and influence between the robots in the robot array, and performs collaborative planning and allocation of the stress field generated by them to make the stress distribution of the entire robot array more reasonable; the imprinting force dynamic balance control method, based on the result of the robot array stress field collaborative allocation, constructs a method that can control and adjust the imprinting force in real time and dynamically, realizes the collaborative compensation and balance of the array imprinting force, and ensures that the pressure applied by each robot during the imprinting process is uniform and consistent. The paper moisture and heat coupling path scheduling unit is used to construct an adaptive scheduling method for printing paths based on the prediction of paper moisture and heat coupling state, and dynamically adjust the printing sequence and path planning of the robot array. Among them, the paper humidity and heat coupling state prediction comprehensively considers the influence of temperature and humidity on the paper during the printing process and predicts the state changes of the paper under different humidity and heat conditions, such as the expansion, contraction and deformation of the paper; the printing path adaptive scheduling method, based on the paper humidity and heat coupling state prediction results, constructs a method that can automatically and flexibly adjust the printing sequence and path planning of the robot array, dynamically adapt to the state of the paper under different humidity and heat conditions, and ensure the printing quality. The special paper texture full-dimensional adjustment unit is used to achieve full-dimensional dynamic adjustment of the special paper texture imprinting robot array through the synergistic effect of the above three mechanisms, so as to complete the differentiated and consistent texture imprinting of special paper. Among them, the full-dimensional dynamic adjustment, through the synergistic effect of three mechanisms—multi-scale morphology perception imprint control unit, robot array stress equalization control unit, and paper wet and heat coupling path scheduling unit—comprehensively and dynamically adjusts the special paper texture imprint robot array from multiple dimensions, such as imprint depth, pressure, stress distribution, and imprint path, to achieve differentiated and consistent texture imprinting of special paper.

[0023] It should be noted that during use, the multi-scale morphology perception imprint control unit can accurately construct a predictive adjustment mechanism based on multi-scale surface morphology, dynamically control the imprint depth and pressure, and ensure that the texture imprint meets expectations. The robot array stress equalization control unit constructs an equalization control method through stress field collaborative distribution, realizes collaborative compensation of imprint force, and avoids imprint defects caused by uneven local stress. The paper moisture and heat coupling path scheduling unit plans the imprint path based on the predicted paper moisture and heat coupling state, and can dynamically adjust the sequence to adapt to changes in paper state and improve imprint quality. The special paper texture full-dimensional adjustment unit integrates the above mechanisms to achieve full-dimensional dynamic adjustment, which can not only meet the differentiated texture requirements of special paper, but also ensure texture consistency. The overall design provides precise control from multiple dimensions, improves production quality and efficiency, and enhances system adaptability and flexibility.

[0024] In one embodiment, a texture imprinting depth prediction and adjustment mechanism is constructed based on multi-scale surface topography perception, specifically including: A multi-scale structured light scanning module is deployed at the front end of the imprinting robot array. A non-contact scanning method is used to perform a micron-level three-dimensional topographic scan of the special paper surface, covering the entire area of ​​the paper imprinting operation, and accurately obtaining the fiber distribution density at various locations on the paper surface. Microporous structure pore size distribution Initial texture roughness parameters ,in These are two-dimensional planar coordinates on the paper surface, representing the specific positional information of the paper. The three-dimensional morphology parameters are preprocessed and fused using a morphology spectrum feature extraction algorithm. First, the parameters are normalized to eliminate dimensional differences. Then, based on the characteristics of the special paper material and the requirements of the printing process, corresponding weight coefficients are assigned to each parameter to construct a paper texture printing plasticity index. This index is a value between 0 and 1. The larger the value, the stronger the embossing deformation ability of the paper in the corresponding area. It is specifically used to characterize the embossing deformation ability of paper in different areas. The calculation formula is: in, The feature weight coefficients are given, and satisfy the following conditions: The weighting coefficients are determined based on the material type of the specialty paper, such as plant fiber paper, synthetic fiber paper, and composite specialty paper, as well as the fineness level of the target embossing texture. They are determined through multiple sets of single-factor variable embossing experiments and grey relational analysis. The parameters with a higher influence on the embossing deformation ability are assigned a larger weighting coefficient. coordinates fiber distribution density, These represent the maximum and minimum values ​​of fiber distribution density on the entire paper surface, respectively. coordinates The average pore size of the microporous structure at that location is determined by the micropore size distribution at that location. The average value is obtained. These represent the maximum and minimum values ​​of the average pore size of the micropores on the entire sheet of paper, respectively. coordinates Initial texture roughness at the location, These are the maximum and minimum values ​​of the initial texture roughness of the entire paper sheet, respectively. A model for predicting imprint depth is constructed. The model adopts a lightweight convolutional neural network (CNN) with a fully connected layer. The model training steps are as follows: Dataset construction: collect surface morphology parameters of at least 50 different types of specialty paper, including: fiber distribution density, average micropore diameter, initial texture roughness, PEI index and texture forming quality data under the corresponding imprinting depth-pressure combination. The dataset is divided into training set, validation set and test set in a ratio of 7:2:1. The model is initialized by setting the CNN layer to contain 3 convolutional blocks. Each convolutional block consists of a convolutional layer with a kernel size of 3×3 and a stride of 1, a batch normalization layer, a ReLU activation function, and a max pooling layer with a kernel size of 2×2. The fully connected layer contains 2 hidden layers with 128 and 64 neurons respectively. The output layer has 2 neurons, corresponding to the imprinting depth and imprinting pressure respectively. For model training, the optimizer was set to Adam, the initial learning rate was 0.001, and a cosine annealing strategy was used to dynamically adjust the learning rate. The batch size was 32, the number of training epochs was 100, and the mean squared error (MSE) was used as the loss function. The loss function formula is as follows:

[0025] in, These are the actual optimal imprinting parameters. These are the model's predicted values. This refers to the batch sample size. Model validation and optimization: The model accuracy is evaluated using a validation set every 10 epochs of training. Training is stopped when the validation set loss does not decrease for 15 consecutive epochs. L2 regularization is used with a regularization coefficient of 0.0001 to prevent overfitting. The final model's prediction accuracy is no less than 95%. Imprint the plasticity index of the texture in different parts of the paper. Using the core input features of the model, combined with process parameters such as the material type of the specialty paper and the specifications of the target embossing texture, the optimal embossing depth at each location on the paper is calculated in real time. With imprinting pressure The combined parameters are calculated using the following formula:

[0026] in, The reference imprint depth is determined based on the target texture. The reference imprint pressure corresponds to the reference imprint depth. This is the adjustment coefficient for the imprinting parameters. Based on the material type and texture embossing process requirements of specialty paper, the embossing texture forming quality scoring model was calibrated through multiple sets of orthogonal experiments. The scoring model takes texture clarity and edge regularity as the core indicators. The calculated optimal imprint depth With imprinting pressure To control the instructions, the motion control system and pressure regulation system of the robot array control each printing robot to dynamically adjust the stroke distance and output pressure of the printing head in different areas of the paper, so as to achieve precise control of the regional differential printing depth of the paper and match the actual deformation capacity of the paper in each area.

[0027] This design employs a multi-scale structured light scanning module using line laser structured light scanning, which can scan the surface of specialty paper in real time, without contact, and over the entire area to obtain precise three-dimensional morphological parameters. By constructing a plasticity index through a morphological spectrum feature extraction algorithm and combining it with a lightweight convolutional neural network to build a prediction model, the optimal imprinting parameters can be calculated in real time. This allows for precise understanding of paper characteristics and dynamic adjustment of imprinting depth and pressure based on the actual conditions of different areas, achieving differentiated and precise imprinting. This avoids uneven imprinting caused by different characteristics of different areas of the paper, effectively improving imprinting quality and meeting the needs of differentiated and consistent texture imprinting for specialty papers.

[0028] In one embodiment, a dynamic equilibrium control method for imprinting force based on the collaborative allocation of stress field of robot array is constructed, specifically including: A six-dimensional force sensing unit is installed between the end effector and the printing head of each printing robot. The sensing unit communicates with the robot control system in real time, and collects the contact force at the contact point between the printing head and the paper at a preset sampling frequency throughout the entire printing process. With micro-vibration acceleration ,in A unique identifier for the robot array, used to distinguish different embossing robots. This refers to the timeframe for the embossing process; The collected contact force and micro-vibration raw information are filtered and denoised to remove invalid data caused by interference factors such as environmental vibration and equipment operation, so as to obtain the effective contact force that can truly reflect the imprinting contact state. The filtering formula is:

[0029] in, The length of the filter window is set according to the sampling frequency and is a positive integer; The sampling time interval of the sensing unit; The attenuation coefficient is determined based on the vibration characteristics of the equipment and the speed level of the embossing operation. The faster the embossing speed, the lower the attenuation coefficient. The smaller the value, the higher the weight of the current sampled data. A stress field distribution model for the imprinting region was constructed. The model adopted a Gaussian process regression (GPR) model, and the model training steps were as follows: Dataset construction involves collecting measured data on contact force, paper tension, imprint position, and corresponding stress distribution under different robot array arrangements and different special paper materials. The dataset size is no less than 10,000 sets, which are divided into training set and test set in an 8:2 ratio. Model initialization: Set the kernel function to a quadratic exponential kernel function. The kernel function formula is as follows:

[0030] in, For signal variance, For length scale, For noise variance, It is the Dirac function; Model training employs maximum likelihood estimation to optimize kernel function hyperparameters. The optimization objective is to maximize the log marginal likelihood. The number of iterations is no less than 500, and the convergence threshold is 1e-6. Model validation: Coefficient of determination between predicted and measured stress distribution values ​​on the test set. Not less than 0.98; Based on the effective contact force of each robot Based on the data source, combined with real-time detection of paper tension status. and the imprinting position coordinates of each robot imprinting head A trained Gaussian process regression model was used to perform global interpolation calculations on the discrete robot imprinting point stress values, resulting in a continuous overall imprinting stress distribution across the entire imprinting area of ​​the paper. The formula accurately characterizes the stress state at various locations on the paper during the printing process.

[0031] in, This refers to the number of robots in the robot array that participate in the embossing operation. For the first The actual contact area between the robot's printing head and the paper is adjusted according to the specifications of the printing head and the pressure. The bandwidth parameter of the Gaussian kernel function is calibrated based on the spacing of the robot array and the effective working range of the imprinting head. A larger spacing results in... The larger the value; The printing stress threshold is preset according to the material strength of the specialty paper and the requirements of the printing process. This threshold is determined through a paper embossing failure limit test based on the tensile strength, tear strength, and other mechanical properties of special paper. It represents the optimal stress value that the paper can withstand without embossing failure or texture distortion, and is set at 70%-85% of the critical stress for paper embossing failure. The embossing stress distribution is calculated in real time. With stress threshold Perform position-by-position comparison; When the imprinting stress in a certain area If the pressure is too high, it is likely to cause paper damage or excessively deep texture. when If the pressure is insufficient, it can easily lead to unclear texture imprinting and poor molding effect. An array-based collaborative adjustment algorithm is designed based on a distributed collaborative control algorithm to adjust the stress field distribution in the imprinting region. As input to the algorithm, the abnormal pressure area and adjacent imprinting robots are identified. The adjacent robots are then controlled to perform real-time force compensation adjustments based on the stress field distribution. The pressure adjustment amount of the adjacent robots is... The calculation formula is:

[0032] in, Number the adjacent robots for which pressure adjustment is to be performed; It is the stress compensation coefficient, and , The response accuracy of the robot array is determined based on the softness of the special paper and the responsiveness of the material; the softer the material and the higher the response accuracy, the better. The closer the value is to 1, the better it controls the range of pressure adjustment, thus avoiding secondary pressure abnormalities caused by excessive adjustment. Based on the calculated pressure adjustment amount The output printing pressure of each robot is updated in real time to form a dynamic pressure balancing mechanism for the robot array, so as to achieve uniform distribution of stress across the entire paper during the printing process and achieve a coordinated control effect of continuous and consistent printing texture.

[0033] This design, with a six-dimensional force sensing unit installed at the robot's end effector, can collect contact force and micro-vibration information in real time. After filtering, a stress field distribution model is constructed. By comparing it with the stress threshold, a distributed collaborative control algorithm is used to achieve real-time force compensation and adjustment in areas of abnormal pressure. It can sense the stress state of each position on the paper in real time during the printing process, adjust the pressure in time, and avoid paper damage or unclear texture caused by excessive or insufficient local pressure. This achieves uniform stress distribution across the entire paper during the printing process, ensuring continuous and consistent printing texture and improving the printing effect.

[0034] In one embodiment, an adaptive scheduling method for the printing path is constructed based on the prediction of the paper's wet and hot coupling state, specifically including: A humidity sensor array and an infrared temperature detection module are deployed in the pre-pressing section of the special paper conveying channel. The detection area of ​​the sensors and detection module covers the entire width of the paper conveying process. The paper surface is scanned position by position at preset fixed sampling intervals to obtain the moisture content of different areas of the paper in real time. With surface temperature ,in These are two-dimensional planar coordinates on the paper surface, representing the specific positional information of the paper. A paper moisture-heat coupled deformation prediction model was established. The model adopted a Long Short-Term Memory (LSTM) network, and the model training steps were as follows: Dataset construction involved collecting measured data on the moisture content, surface temperature, and corresponding linear expansion rate of specialty papers under different temperature and humidity conditions, covering a temperature range of 15-35℃ and a humidity range of 30%-80%. The dataset is divided into training, validation, and test sets in a 7:2:1 ratio;

[0035] Model initialization: The LSTM layer is set to contain 2 hidden layers, with 64 neurons in each layer. The input layer dimension is 2 (water content and surface temperature), the output layer dimension is 1 (linear dilatation rate), and the forget gate bias is initially set to 1.0. For model training, the optimizer used was RMSprop, the learning rate was 0.0005, the batch size was 64, the number of training epochs was 200, and the loss function was the mean absolute error (MAE). The loss function formula is as follows:

[0036] An early stopping strategy with a patience of 20 is used to prevent overfitting. Model validation showed that the mean absolute percentage error (MAPE) between the predicted and measured linear expansion rates on the test set did not exceed 2%. Based on the trained LSTM model and considering the fiber material characteristics of specialty paper, the local linear expansion rate of paper caused by moisture and heat during the printing process is predicted. The expansion rate is positive if the area of ​​the paper expands, and negative if the area shrinks. The calculation formula is:

[0037] in, This is the coefficient of thermal expansion of paper. The coefficient of linear expansion due to humidity is determined based on the material type of special paper through a constant temperature and humidity change experiment and a constant humidity and temperature change experiment. The experimental environment simulates the temperature and humidity range of the actual printing workshop. The standard room temperature required for the embossing process. The moisture content of specialty paper is the standard mass fraction, and all values ​​are preset values ​​for the process. Based on the printing process requirements and paper deformation characteristics, two linear expansion rate thresholds are set. and And satisfy , The high humidity expansion threshold is determined based on the maximum expansion rate of specialty paper without texture deformation under standard printing processes, and is set at 0.05% to 0.2%. The drying shrinkage threshold is determined based on the maximum shrinkage rate of specialty paper without texture cracking under standard printing processes, and is ranged from -0.2% to 0.05%. The predicted local linear expansion rate By comparing with a threshold, the paper imprint area type is classified: when When the area is identified as a high humidity expansion zone, the paper in this area is prone to deformation due to its high moisture content, so it is necessary to press it in advance and reduce the pressure. when When the paper is in this area, it is determined to be a drying shrinkage zone. Due to its low moisture content and high hardness, the printing process in this area needs to be delayed and the pressure increased. when When the area is identified as a normal area, the paper in this area is in a stable state of moisture and heat, and printing is performed according to the reference path. For the different types of imprinting areas after division, the imprinting sequence and imprinting path planning of the robot array are dynamically adjusted through path planning algorithms, taking into account the robot array's layout and operating efficiency. Imprinting operations are prioritized for high-humidity expansion areas, while imprinting operations for drying and shrinking areas are delayed. At the same time, corresponding imprinting pressure correction values ​​are matched for different areas. To achieve adaptation to pressure and humid / heat conditions, the calculation formula is as follows:

[0038] in, This is a humidity and pressure correction factor, and , The calibration is based on the moisture and heat deformation sensitivity of specialty paper and the forming requirements of the target texture. Higher moisture and heat deformation sensitivity and finer texture result in... The larger the value; The reference imprinting pressure; Based on the adjusted printing sequence, path planning, and printing pressure, the robot array scheduling system and motion control system control each printing robot to perform printing operations according to the new plan, thereby achieving adaptive scheduling control of the printing path and avoiding problems such as texture deformation and misalignment of paper caused by changes in humidity and temperature at the process path level.

[0039] This design deploys a humidity sensor array and an infrared temperature detection module to acquire paper moisture and heat data, establishes a moisture and heat coupled deformation prediction model to predict the linear expansion rate, classifies the imprinting area type, and dynamically adjusts the imprinting sequence and path planning. It can predict the deformation of paper caused by moisture and heat changes in advance, avoid problems such as texture deformation and misalignment from the process path level, prioritize the treatment of easily deformable areas, match the corresponding imprinting pressure, make the imprinting operation more in line with the actual state of the paper, improve imprinting quality, and reduce imprinting defects caused by moisture and heat factors.

[0040] In one embodiment, the multi-scale structured light scanning module employs line laser structured light scanning, with a scanning accuracy of [missing information]. The scan frame rate is not lower than The scanning resolution is adjustable according to the paper imprinting accuracy requirements, enabling real-time, non-contact, full-area three-dimensional scanning of the surface of special paper, and the scanning data can be transmitted to the robot control system in real time for subsequent processing.

[0041] This design enables real-time, non-contact, full-area three-dimensional scanning of the surface of specialty paper, and the scanned data can be transmitted to the control system in real time. High-precision scanning can accurately acquire the details of the paper surface, providing an accurate basis for subsequent printing depth prediction and adjustment. Real-time data transmission ensures that the system responds in a timely manner and dynamically adjusts the printing parameters according to the paper surface conditions, improving the timeliness and accuracy of printing and meeting the requirements of high-precision printing.

[0042] In one embodiment, the six-dimensional force sensing unit adopts a high-precision piezoelectric sensing structure, with a force detection accuracy of 0.01N, a detectable force range of 0~500N, and a micro-vibration acceleration detection range of -10g~+10g. The acceleration detection accuracy is 0.001g, and the sampling frequency can be adjusted according to the printing speed. It can capture the changes in contact force and micro-vibration during the printing process in real time and accurately, and the detection data is transmitted to the collaborative control system of the robot array without delay.

[0043] This design allows for real-time and precise capture of changes in contact force and micro-vibrations during the printing process. Precise force detection ensures accurate perception of printing stress, providing reliable data for stress balance control. The adjustable sampling frequency adapts to different printing speeds, and data transmission without delay ensures timely pressure adjustment by the system, achieving dynamic balance control of printing force and improving printing quality.

[0044] In one embodiment, the humidity sensor employs a capacitive humidity sensing chip with a detection accuracy of [missing information]. The detection range is The infrared temperature detection module uses a non-contact infrared temperature probe with a temperature detection accuracy of ±0.1℃ and a detection range of 0~80℃. The sampling interval between the sensor and the detection module is 5~50ms. The sampling interval is determined based on the paper conveying speed and the working width of the printing head. The faster the conveying speed and the larger the working width, the smaller the sampling interval value. It can be adaptively adjusted according to the paper conveying speed. The detection data is uploaded to the paper status analysis system in real time to meet the real-time and accurate detection requirements of the paper's damp and hot state.

[0045] This design meets the need for real-time and accurate detection of the paper's moisture and heat status. High-precision detection can accurately obtain the paper's moisture content and surface temperature, providing reliable data for predicting moisture-heat coupled deformation. The adaptive sampling interval can be adjusted according to the paper's conveying speed, ensuring the timeliness and completeness of the detection data. This allows the system to dynamically adjust the printing path based on accurate moisture and heat data, avoiding printing problems caused by moisture and heat factors.

[0046] In one embodiment, the system further includes a closed-loop feedback correction step. After the printing operation is completed, a high-precision texture detection device is used to perform full-area detection on the surface texture of the printed paper to obtain parameters such as the depth, clarity, and continuity of the actual texture. The actual detection results are compared with the target texture parameters position by position to calculate the texture deviation value. The threshold for determining the texture deviation value is determined based on the accuracy level of the target embossed texture. The deviation threshold for fine texture is no greater than 0.05mm, and the deviation threshold for regular texture is no greater than 0.1mm. Based on the magnitude and distribution of the texture deviation value, combined with the previous embossing parameter adjustment data, the adjustment coefficients and threshold parameters of embossing depth, embossing pressure, and embossing path are updated through iterative optimization algorithms. The updated parameters are then fed back to the control systems of the robot array to achieve closed-loop dynamic adjustment of embossing parameters and continuously improve the forming accuracy of the embossed texture.

[0047] This design uses high-precision texture detection equipment to detect the texture of the printed paper, calculate the deviation value, update the printing parameter adjustment coefficient and threshold parameters based on the deviation, and feed the feedback to the control system. It can promptly detect deviations between the printed texture and the target parameters, and use iterative optimization algorithms to dynamically adjust the printing parameters in a closed loop. Continuous parameter updates enable the system to adapt to changes in paper characteristics and the printing environment, continuously improve the forming accuracy of the printed texture, ensure high quality for each printing, reduce scrap rate, and improve production efficiency.

[0048] Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0049] The above embodiments provide a detailed description of the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A dynamic adjustment system for a special paper texture imprinting robot array, characterized in that, The system includes: Multi-scale morphology perception and imprinting control unit, robot array stress equalization control unit, paper wet and heat coupling path scheduling unit, and special paper texture full-dimensional adjustment unit. The multi-scale topography perception imprint control unit constructs a texture imprint depth prediction and adjustment mechanism based on multi-scale surface topography perception, which is used to dynamically control the imprint depth and output pressure of the robot array. The robot array stress equalization control unit constructs a dynamic equalization control method for imprinting force based on the collaborative allocation of robot array stress field, thereby realizing real-time collaborative compensation and global equalization regulation of array imprinting force. The paper moisture and heat coupling path scheduling unit constructs an adaptive scheduling method for printing paths based on the prediction of paper moisture and heat coupling state, and dynamically adjusts the printing operation sequence and path planning strategy of the robot array. The special paper texture full-dimensional adjustment unit, in coordination with the operation mechanism and control data of the above three units, realizes full-dimensional dynamic adjustment of the special paper texture imprinting robot array, completes differentiated and consistent texture imprinting operations of special paper, and adapts to the imprinting process requirements and target texture specifications of different types of special paper.

2. The special paper texture imprinting robot array dynamic adjustment system according to claim 1, characterized in that, The multi-scale topography sensing imprint control unit specifically includes the following steps: Equipped with a non-contact multi-scale structured light scanning module, the multi-scale structured light scanning module is deployed at the front end of the imprinting robot array to perform micron-level three-dimensional morphology full-domain scanning on the surface of special paper, covering the entire area of ​​the paper imprinting operation; Accurately acquiring three-dimensional morphological parameters such as fiber distribution density, micropore size distribution, and initial texture roughness at various locations on the paper surface provides comprehensive and high-precision basic data support for the dynamic control of printing depth and pressure.

3. The special paper texture imprinting robot array dynamic adjustment system according to claim 2, characterized in that, The multi-scale topography sensing imprint control unit further includes the following steps: Built-in morphology spectrum feature extraction algorithm and imprint depth prediction model; The morphology spectrum feature extraction algorithm performs normalization preprocessing and multi-dimensional feature fusion on the three-dimensional morphology parameters. It assigns corresponding weight coefficients to each parameter in combination with the characteristics of special paper material and the requirements of printing process, and constructs a texture printing plasticity index that characterizes the printing deformation ability of different areas of paper. The imprint depth prediction model uses the texture imprint plasticity index as the core input feature, and combines the material type of the special paper and the process parameters of the target imprint texture fineness level to calculate the optimal imprint depth and imprint pressure combination parameters at each position of the paper in real time.

4. The special paper texture imprinting robot array dynamic adjustment system according to claim 3, characterized in that, The multi-scale topography sensing imprint control unit includes the following steps: Establish real-time communication connections with the motion control system and pressure regulation system of the robot array; Using the calculated optimal combination of printing depth and printing pressure as control commands, each printing robot dynamically adjusts the travel distance and output pressure of the printing head in different areas of the paper, achieving precise control of the regionally differentiated printing depth of different positions on the paper, matching the actual printing deformation capacity of each area of ​​the paper, and ensuring the forming quality of the texture printing.

5. The special paper texture imprinting robot array dynamic adjustment system according to claim 1, characterized in that, The robot array stress equalization control unit includes the following steps: A force sensing unit is configured between the end effector and the impression head of each imprinting robot; The force sensing unit communicates with the robot control system in real time. During the entire printing process, it collects the contact force and micro-vibration acceleration information of each robot printing head and the paper at a preset sampling frequency, providing real-time detection data for the balanced control of printing force.

6. The special paper texture imprinting robot array dynamic adjustment system according to claim 5, characterized in that, The robot array stress equalization control unit further includes the following steps: Built-in filtering and noise reduction algorithm and stress field distribution model of the imprinting area; The filtering and denoising algorithm filters the collected raw information of contact force and micro-vibration acceleration, eliminating invalid data caused by interference factors such as environmental vibration and equipment operation, and obtaining effective contact force data that can truly reflect the imprinting contact state. The stress field distribution model of the imprinting area uses effective contact force as the basic data source. It combines the real-time detected paper tension state and the imprinting position coordinates of each robot imprinting head to calculate the continuous overall imprinting stress distribution in the entire imprinting area of ​​the paper through global interpolation, accurately representing the real-time stress state of each position of the paper during the imprinting process.

7. The special paper texture imprinting robot array dynamic adjustment system according to claim 6, characterized in that: The robot array stress equalization control unit presets the printing stress threshold according to the material strength of the special paper and the printing process requirements. It compares the real-time calculated printing stress distribution with the printing stress threshold position by position to accurately determine abnormal areas where the printing pressure is too high or too low. The unit also incorporates a distributed collaborative control algorithm. Using the stress field distribution in the printing area as the algorithm input, it identifies areas of abnormal pressure and adjacent printing robots. It controls adjacent robots to perform real-time force compensation adjustments based on the stress field distribution, calculates the corresponding pressure adjustment amount, and updates the output printing pressure of each robot in real time, forming a dynamic pressure balancing mechanism for the robot array to achieve uniform distribution of stress across the entire paper during the printing process.

8. The special paper texture imprinting robot array dynamic adjustment system according to claim 1, characterized in that, The paper moisture and heat coupling path scheduling unit includes the following steps: A temperature and humidity detection module is installed in the pre-printing section of the special paper conveying channel; The temperature and humidity detection module includes a humidity sensor array and an infrared temperature detection module, and the detection area covers the entire width of the paper conveying process. The paper surface is scanned and detected position by position according to a preset fixed sampling interval, and the moisture content and surface temperature parameters of different areas of the paper are obtained in real time, so as to realize the full-area and real-time detection of the paper's moisture and heat status.

9. The special paper texture imprinting robot array dynamic adjustment system according to claim 8, characterized in that: The paper moisture-heat coupling path scheduling unit has a built-in moisture-heat coupling deformation prediction model and path planning algorithm. The moisture-heat coupling deformation prediction model takes the paper's moisture-heat state parameters as input, combines the fiber material characteristics of special paper, predicts the local linear expansion rate of paper affected by moisture and heat during the printing process, and classifies the paper printing area into three types based on the preset linear expansion rate threshold: high humidity expansion, drying shrinkage, and normal. The path planning algorithm dynamically adjusts the printing sequence and path planning strategy of the robot array based on the type of printing area, the arrangement of the robot array, and the work efficiency. At the same time, it matches the corresponding printing pressure correction value for different areas to achieve precise matching between printing pressure and paper humidity and heat conditions, thereby avoiding texture deformation and misalignment problems caused by changes in paper temperature and humidity at the process path level.

10. The special paper texture imprinting robot array dynamic adjustment system according to claim 1, characterized in that, It also includes a closed-loop feedback and correction mechanism: The closed-loop feedback correction link is equipped with a high-precision texture detection device. After the printing operation is completed, the texture of the printed paper surface is detected over the entire area to obtain the parameters of the actual texture depth, clarity and continuity. The actual detection results are compared with the target texture parameters position by position to calculate the texture deviation value. The closed-loop feedback correction process also incorporates an iterative optimization algorithm. Based on the magnitude and distribution of the texture deviation value, combined with the previous imprinting parameter adjustment data, it updates the adjustment coefficients and threshold parameters of the imprinting depth, imprinting pressure, and imprinting path. The updated parameters are then fed back to the control systems of the robot array in real time, realizing closed-loop dynamic adjustment of the imprinting parameters, continuously optimizing the imprinting parameter configuration, and improving the forming accuracy and consistency of the imprinted texture.

Images