Viscosity-variable UV ink and intelligent control method thereof

Abstract

The invention discloses a variable-viscosity UV ink system and an intelligent control method thereof, and relates to the technical field of data processing, and the system comprises a multi-dimensional sensing module which is used for executing multi-dimensional sensing data collection and establishing an ink state feature vector; the prediction module is used for outputting predicted viscosity distribution parameters by using the double-domain collaborative modeling unit; the fusion module is used for authenticating fusion and outputting fusion viscosity state parameters; the feedback module is used for performing self-adaptive feedback optimization according to a difference comparison result of the target viscosity curve and the fused viscosity state parameters; and the management module is used for performing intelligent control management according to the self-adaptive feedback optimization result. The technical problems that in the prior art, UV ink viscosity control is single and lagged, the multi-dimensional cooperative regulation and control capacity is lacked, and light curing and rheological property coupling cannot be dealt with are solved, and the technical effects that real-time, accurate and self-adaptive prediction and cooperative control over the ink viscosity are achieved, and therefore the printing quality and the process stability are improved are achieved.

Description

Technical Field

[0001] This application relates to the field of data processing technology, specifically to variable viscosity UV ink and its intelligent control method. Background Technology

[0002] Ultraviolet (UV) inks, as functional materials, are widely used in high-precision manufacturing fields, such as micro / nano structure printing, optical device fabrication, and biomedical sensor manufacturing. However, the viscosity behavior of UV inks during actual printing is affected by various factors, exhibiting highly time-varying and nonlinear characteristics. Factors such as temperature changes, flow shear history, illumination conditions, and differences in formulation components can all lead to uneven curing, tapering, stringing, or a decrease in forming accuracy. Traditional UV ink systems employ local PID control based on temperature or flow rate, which struggles to achieve real-time and precise control of ink viscosity. This is especially true in high-resolution, multi-material composite printing scenarios, where there is a strong coupling between the ink's photocuring process and rheological properties. Single-quantity control cannot adapt to complex dynamic processes and generally lacks the ability to perceive and integrate the ink's state in multiple dimensions, failing to accurately capture key state information such as changes in surface tension and the progress of the curing reaction, resulting in lag in control response and poor adaptability.

[0003] Therefore, current technologies for UV inks suffer from limitations such as single and lagging viscosity control, lack of multi-dimensional synergistic regulation capabilities, and inability to address the technical challenges of coupling photocuring and rheological properties. Summary of the Invention

[0004] This application provides a variable viscosity UV ink and its intelligent control method, which solves the technical problems of existing UV ink viscosity control being singular, lagging, and lacking multi-dimensional collaborative regulation capabilities, and unable to cope with the coupling of photocuring and rheological properties. It achieves the technical effect of realizing real-time accurate, adaptive prediction and collaborative control of ink viscosity, thereby improving printing quality and process stability.

[0005] This application provides a variable viscosity UV ink system, comprising: a multi-dimensional sensing module for performing multi-dimensional sensing data acquisition, the multi-dimensional sensing data including ink temperature, flow rate, surface tension, and photocuring process status, and establishing an ink state feature vector based on the multi-dimensional sensing data; a prediction module for inputting the ink state feature vector into a dual-domain collaborative modeling unit including an optical domain model and a rheological domain model, and outputting predicted viscosity distribution parameters; a fusion module for synchronously acquiring local viscosity data, performing certified fusion based on the local viscosity data and the predicted viscosity distribution parameters, and outputting fused viscosity state parameters; a feedback module for inputting the fused viscosity state parameters into a feedforward control layer, performing adaptive feedback optimization based on the difference comparison result between the target viscosity curve and the fused viscosity state parameters, and establishing an adaptive feedback optimization result; and a management module for performing intelligent control management based on the adaptive feedback optimization result, the adaptive feedback optimization result including light source power correction value, nozzle heating adjustment value, and pulse timing correction value.

[0006] In a possible implementation, the feedback module inputs the fusion viscosity state parameter to the feedforward control layer, including: reading the printer's printing target data; analyzing the target deposition path features based on the printing target data, whereby the target deposition path features include spraying speed features, linewidth features, and layer thickness requirement features; sending the target deposition path features to the feedforward control layer; fitting the viscosity requirement using the fitting channel of the feedforward control layer to establish a target viscosity curve; performing node-by-node difference analysis using the target viscosity curve and the fusion viscosity state parameter to establish a difference comparison result; and using the difference comparison result for adaptive feedback optimization.

[0007] In a possible implementation, the variable viscosity UV ink system further includes: acquiring historical control response data of the printer; configuring a response accuracy confidence index based on the historical control response data; performing node-by-node confidence analysis of the target viscosity curve using the response accuracy confidence index to establish node confidence labels; configuring node hysteresis influence factors using the node confidence labels; and performing adaptive feedback optimization of sequential nodes based on the node hysteresis influence factors and difference comparison results.

[0008] In a possible implementation, the variable viscosity UV ink system further includes: configuring a hysteresis cumulative impact term for each node based on the node hysteresis impact factor and the difference comparison result; establishing a smoothing cost term between nodes; establishing a node objective function based on the hysteresis cumulative impact term, the smoothing cost term, and the control energy cost term; and using the node objective function to perform adaptive feedback optimization management for each node.

[0009] In a possible implementation, the variable viscosity UV ink system further includes: parsing the ink state feature vector to obtain light source power, wavelength, pulse mode feature components, and ink temperature, flow rate, surface tension, and component concentration; inputting the light source power, wavelength, and pulse mode feature components into an optical domain model to output the photoinitiator activation rate; and after evaluating the molecular chain entanglement based on the ink temperature, flow rate, surface tension, and component concentration, inputting the molecular chain entanglement and photoinitiator activation rate into a rheological domain model to output predicted viscosity distribution parameters.

[0010] In a possible implementation, the variable viscosity UV ink system further includes: acquiring the hysteresis confidence of local viscosity data; acquiring the prediction uncertainty of the predicted viscosity distribution parameters; and performing weighted fusion of the local viscosity data and the predicted viscosity distribution parameters based on the hysteresis confidence and the prediction uncertainty to establish fused viscosity state parameters.

[0011] In a possible implementation, the variable viscosity UV ink system further includes: a verification module, used to establish verification nodes based on the adaptive feedback optimization results, and to perform control verification based on the target viscosity curve using the verification nodes to establish verification feedback; and a feedback module, used to establish sequence compensation based on the verification feedback, and to perform node-by-node feedback optimization of the adaptive feedback optimization results using the sequence compensation.

[0012] In a possible implementation, the variable viscosity UV ink system further includes: an early warning module, used to identify deviation threshold triggers in the verification feedback, establish deviation threshold trigger identification results, and manage control anomaly reporting based on the deviation threshold trigger identification results.

[0013] In a possible implementation, the variable viscosity UV ink system further includes: the fusion module includes a multi-point local viscosity sensor acquisition unit, which is used to perform weighted authentication on multiple local viscosity data in the same area, and output the redundant weighted authentication results as local viscosity data.

[0014] This application also provides an intelligent control method for variable viscosity UV ink, the method comprising: performing multi-dimensional sensing data acquisition, the multi-dimensional sensing data including ink temperature, flow rate, surface tension, and photocuring process status; establishing an ink state feature vector based on the multi-dimensional sensing data; inputting the ink state feature vector into a dual-domain collaborative modeling unit including an optical domain model and a rheological domain model, and outputting predicted viscosity distribution parameters; synchronously acquiring local viscosity data, performing authentication fusion based on the local viscosity data and the predicted viscosity distribution parameters, and outputting fused viscosity state parameters; inputting the fused viscosity state parameters into a feedforward control layer, performing adaptive feedback optimization based on the difference comparison result between the target viscosity curve and the fused viscosity state parameters, and establishing an adaptive feedback optimization result; and performing intelligent control management based on the adaptive feedback optimization result, the adaptive feedback optimization result including light source power correction value, nozzle heating adjustment value, and pulse timing correction value.

[0015] This application proposes a variable viscosity UV ink and its intelligent control method, comprising a multi-dimensional sensing module for acquiring multi-dimensional sensing data and establishing an ink state feature vector; a prediction module for outputting predicted viscosity distribution parameters using a dual-domain collaborative modeling unit; a fusion module for verifying and fusion output of fused viscosity state parameters; a feedback module for adaptive feedback optimization based on the difference comparison results between the target viscosity curve and the fused viscosity state parameters; and a management module for intelligent control management based on the adaptive feedback optimization results. This addresses the technical problems of existing UV ink viscosity control technologies, which suffer from single, lagging, and lacking multi-dimensional collaborative regulation capabilities, and are unable to handle the coupling of photocuring and rheological properties. It achieves real-time, accurate, adaptive prediction, and collaborative control of ink viscosity, thereby improving printing quality and process stability. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings of the embodiments of this disclosure will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0017] Figure 1 This is a schematic diagram of the variable viscosity UV ink system provided in an embodiment of this application.

[0018] Figure 2 This is a schematic diagram of the intelligent control method for variable viscosity UV ink provided in an embodiment of this application.

[0019] Figure labeling: Multi-dimensional perception module 10, prediction module 20, fusion module 30, feedback module 40, management module 50. Detailed Implementation

[0020] The above description is merely an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below.

[0021] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application will be provided in conjunction with the accompanying drawings. The described embodiments should not be considered as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] In the following description, references to "some embodiments" describe a subset of all possible embodiments. However, it is understood that "some embodiments" can be the same or different subsets of all possible embodiments and can be combined with each other without conflict. The terms "first" and "second" are used merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, system, product, or server that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, products, or devices. 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 application belongs. The terminology used herein is for the purpose of describing embodiments of this application only.

[0023] This application provides a variable viscosity UV ink system, such as... Figure 1 As shown, the system includes: The multidimensional sensing module 10 is used to perform multidimensional sensing data acquisition, including ink temperature, flow rate, surface tension, and photocuring process status, and to establish an ink state feature vector based on the multidimensional sensing data.

[0024] Preferably, the multidimensional sensing module is a unit integrating multiple sensors, signal processing circuits, and data integration components. It is used to synchronously collect key parameters that comprehensively describe the instantaneous physical and chemical state of UV ink, i.e., to perform multidimensional sensing data acquisition. Specifically, this includes using a miniature embedded thermocouple or platinum resistance temperature sensor to directly contact the ink flow path and then obtaining the temperature value through analog-to-digital conversion; measuring the ink flow path through a thermal or differential pressure microflow sensor and outputting an electrical signal proportional to the flow rate to collect flow rate data; using a pendant drop optical sensor or a vibration probe sensor near the printhead nozzle to indirectly calculate the surface tension value by analyzing the droplet morphology or vibration frequency; monitoring changes in the absorbance or fluorescence characteristics of the ink in the irradiated area in real time through a miniature online spectrometer or a photodiode of a specific wavelength, for example, monitoring the absorption attenuation of the photoinitiator at a specific wavelength or monitoring the increase in the fluorescence intensity of the product to obtain the state of the photocuring process; and then integrating the ink temperature, flow rate, surface tension, and photocuring process state to form multidimensional sensing data. Then, the multidimensional sensing data is preprocessed, including filtering, calibration, and timestamp alignment. Next, the processed physical quantities are normalized or standardized, for example, mapping all values ​​to the [0, 1] interval or converting them into standard scores with a mean of 0 and a variance of 1 to eliminate dimensional differences. Finally, the state values ​​of different dimensions of the ink at the current moment are constructed in sequence into an ink state feature vector to describe the current comprehensive state of the ink.

[0025] The prediction module 20 is used to input the ink state feature vector into a dual-domain collaborative modeling unit containing an optical domain model and a rheological domain model, and output predicted viscosity distribution parameters.

[0026] Preferably, the prediction module receives the ink state feature vector and uses an embedded dual-domain collaborative modeling unit to calculate the predicted value of ink viscosity over a future period or spatial range. The dual-domain collaborative modeling unit includes an optical domain model and a rheological domain model. The optical domain model is a photochemical reaction kinetic model that, based on the input light source parameters and ink composition, quantitatively calculates the activation rate of the photoinitiator or the conversion rate of the monomer in the ink under the current illumination conditions by solving differential equations, representing the intensity of the photochemical reaction. The rheological domain model is used to evaluate the initial viscosity and molecular chain entanglement of the ink in its uncured or partially cured state based on ink temperature and flow rate. Specifically, the prediction module analyzes the ink state feature vector, decomposing it to obtain the light source power, wavelength, pulse mode feature components, and ink temperature, flow rate, surface tension, and component concentration, which are used as inputs to the optical domain model and the rheological domain model, respectively. Simultaneously, the photoinitiator activation rate output from the optical domain model is input to the rheological domain model, and the predicted viscosity value is output. Due to the inherent uncertainties in the inputs and the reaction itself, the dual-domain collaborative modeling unit ultimately outputs a predicted viscosity distribution parameter, representing the most probable value of the ink viscosity and its uncertainty range.

[0027] Furthermore, the specific configuration of the prediction module 20 also includes: parsing the ink state feature vector to obtain the light source power, wavelength, pulse mode feature components, and ink temperature, flow rate, surface tension, and component concentration; inputting the light source power, wavelength, and pulse mode feature components into the optical domain model to output the photoinitiator activation rate; evaluating the molecular chain entanglement degree based on the ink temperature, flow rate, surface tension, and component concentration, and then inputting the molecular chain entanglement degree and photoinitiator activation rate into the rheological domain model to output the predicted viscosity distribution parameters.

[0028] Preferably, the prediction module parses the received ink state feature vector, including decomposing the ink state feature vector into a light source parameter group and an ink parameter group according to a predefined index position or data tag. The light source parameter group includes physical parameters that control the light source, namely, light source power, wavelength, and pulse mode feature components. The ink parameter group includes the physicochemical state parameters of the ink itself, namely, ink temperature, flow rate, surface tension, and component concentration, such as monomers, oligomers, photoinitiators, pigments, etc.

[0029] Preferably, the light source power, wavelength, and pulse mode characteristic components are input into the optical domain model, and the interaction between light and ink is simulated through mathematical equations. Specifically, the effective instantaneous light intensity acting on a unit volume of ink is calculated based on the light source power and pulse mode; the absorption rate of ink to light of a specific wavelength is calculated using the Beer-Lambert law based on the wavelength and photoinitiator concentration, where the photoinitiator concentration is obtained from the component concentration; then the effective instantaneous light intensity and absorption rate are substituted into a system of ordinary differential equations as inputs to solve for the photoinitiator activation rate, such as the concentration of free radicals generated per unit time or the decomposition rate of the photoinitiator, where the system of ordinary differential equations is used to describe the rate at which the photoinitiator decomposes after absorbing photons to generate free radicals.

[0030] Preferably, the fluid state is preprocessed and calculated for ink temperature, flow rate, surface tension, and component concentration, and the molecular chain entanglement degree is output, representing the initial entanglement degree under the current state, which is used to describe the degree of entanglement and hooking of polymer molecular chains in the solution. Then, the molecular chain entanglement degree and photoinitiator activation rate are input into the rheological domain model. Specifically, the molecular chain entanglement degree is used to set the initial viscosity baseline of the uncured ink, and the photoinitiator activation rate output by the photodomain model is used as the core input variable. According to classical rheological relationships, viscosity is strongly positively correlated with weight-average molecular weight and entanglement degree. Then, the rheological domain model calculates the viscosity that increases sharply due to the increase in molecular weight in real time through equations. Due to the errors in the input parameters and the model itself, uncertainty propagation analysis is performed, and finally, the predicted viscosity distribution parameters are output.

[0031] The fusion module 30 is used to synchronously collect local viscosity data, perform authentication fusion based on the local viscosity data and predicted viscosity distribution parameters, and output fused viscosity state parameters.

[0032] Furthermore, the specific configuration of the fusion module 30 also includes: acquiring the hysteresis confidence of local viscosity data; acquiring the prediction uncertainty of the predicted viscosity distribution parameters; and performing weighted fusion of local viscosity data and predicted viscosity distribution parameters based on the hysteresis confidence and the prediction uncertainty to establish fused viscosity state parameters.

[0033] Preferably, the fusion module is a data processing unit used to synchronously receive predicted viscosity distribution parameters and measured local viscosity data and perform weighted fusion to calculate the current optimal viscosity state estimate. Specifically, a vibratory viscometer is used to synchronously collect local viscosity data. Hysteresis refers to dynamic response characteristics. In particular, time hysteresis refers to the time delay between the change in the physical viscosity of the ink and the output of a stable and accurate reading from the sensor. Hysteresis confidence is an estimated weighting coefficient, which usually ranges from [0, 1]. The lower the weighting coefficient, the lower the confidence in the current local viscosity data. The vibratory viscometer is modeled as a first-order inertial element. When the detected ink state is changing rapidly, the current sensor reading cannot fully represent the true viscosity at the current moment due to the hysteresis effect, and thus the hysteresis confidence is dynamically lowered. Next, the prediction uncertainty of the predicted viscosity distribution parameters is obtained, that is, the variance of the predicted viscosity distribution parameters is calculated to quantify the credibility of the prediction of the dual-domain collaborative modeling unit. The larger the variance, the more uncertain the prediction. Then, the local viscosity data and the predicted viscosity distribution parameters are weighted and fused according to the hysteresis confidence and prediction uncertainty. The weights of the local viscosity data and the predicted viscosity distribution parameters are adaptively determined and sum to 1. That is, when the ink state is stable, the weight of the local viscosity data increases, and when the ink state changes rapidly, the weight of the local viscosity data decreases. Finally, the fused viscosity state parameters are output, and the ink estimation accuracy and response speed during transient changes are ensured.

[0034] Furthermore, the specific configuration of the fusion module 30 also includes a multi-point local viscosity sensor acquisition unit, which is used to perform weighted authentication on multiple local viscosity data in the same area and output the redundant weighted authentication results as local viscosity data.

[0035] Preferably, the multi-point local viscosity sensor acquisition unit is used to perform weighted authentication on multiple local viscosity data in the same area. Specifically, multiple viscosity sensors are installed in the same area very close to the printhead to overcome local deviations caused by differences in sensor manufacturing and local contamination. The weighted authentication includes data validity verification and consistency verification. Then, a weight coefficient is assigned to each sensor reading for weighted averaging. The weight coefficient is configured according to the historical performance of the sensor. Sensors with long-term stability have a higher weight coefficient. Finally, a redundant weighted authentication result is obtained, which is the weighted average of the readings of all effective viscosity sensors. This is used as the local viscosity data output to ensure that the actual viscosity measurement value is more accurate and reliable.

[0036] Feedback module 40 is used to input the fusion viscosity state parameters to the feedforward control layer, and to perform adaptive feedback optimization based on the difference comparison results between the target viscosity curve and the fusion viscosity state parameters, and to establish the adaptive feedback optimization result.

[0037] Furthermore, the specific configuration of the feedback module 40 also includes: reading the printer's printing target data; performing target deposition path feature analysis based on the printing target data, wherein the target deposition path features include spraying speed features, line width features, and layer thickness requirement features; sending the target deposition path features to the feedforward control layer; using the fitting channel of the feedforward control layer to perform viscosity requirement fitting and establish a target viscosity curve; using the target viscosity curve and the fused viscosity state parameters to perform node-by-node difference analysis and establish difference comparison results; and using the difference comparison results to perform adaptive feedback optimization.

[0038] Preferably, the feedback module is an optimization calculation unit used to receive the fused viscosity state parameters and the target viscosity curve, calculate the difference between the two, and then perform adaptive feedback optimization to dynamically adjust the control parameters to achieve high-precision tracking control. Specifically, it reads the printer's printing target data from the host computer instruction file, including the printhead's movement path, such as the X, Y, and Z coordinate sequence, and the amount of ink to be ejected. Then, it performs target deposition path feature analysis on the printing target data to extract key process parameters directly related to fluid behavior, namely, target deposition path features, including spraying speed features, linewidth features, and layer thickness requirement features. Among them, the spraying speed feature refers to the speed at which the printhead moves along the path. The faster the spraying speed, the better the ink flow and the lower the viscosity required. The linewidth feature refers to the width of the ink line to be deposited. The thinner the linewidth, the better the ink's formability, i.e., the higher the viscosity required. The layer thickness requirement feature refers to the thickness of the desired deposited layer, which is related to the nozzle diameter and spraying frequency, and also affects the ink viscosity requirements. The thinner the layer thickness, the more precise the viscosity control required.

[0039] Preferably, the target deposition path features are sent to the feedforward control layer, where the feedforward control layer embeds a fitting channel, which is a mapping relationship between process parameters and ideal viscosity established in advance through a large number of process experiments. Then, the fitting channel of the feedforward control layer is used to fit the viscosity requirements of the target deposition path features. That is, based on the combination of key process parameters contained in the target deposition path features, the mapping relationship between process parameters and ideal viscosity is matched and fitted to calculate the ideal viscosity value of the target ink for all current path nodes, thereby generating a target viscosity value sequence that changes over time, and then plotting the target viscosity curve. Then, a node-by-node difference analysis is performed using the target viscosity curve and the fusion viscosity state parameters. Specifically, the target viscosity value for each node in the current printing path is obtained from the target viscosity curve, and the fusion viscosity state parameter at the current moment is obtained from the fusion viscosity state parameters. The difference is calculated as the difference comparison result, quantifying the gap between the current ink state and the ideal ink state. Next, adaptive feedback optimization is performed using the difference comparison result as input to a feedback optimizer. This feedback optimizer may be a model predictive controller, used to calculate a set of control actions as the adaptive feedback optimization result. These actions include light source power correction values, nozzle heating adjustment values, and pulse timing correction values. For example, increasing the light source power or decreasing the heater temperature ensures that the future ink viscosity prediction output approximates the corresponding target ink viscosity to eliminate errors. This ensures that the ink viscosity changes with the complex variations of the printing path, thereby improving printing quality and process stability.

[0040] Furthermore, the specific configuration of the feedback module 40 also includes: acquiring historical control response data of the printer; configuring a response accuracy confidence index based on the historical control response data; using the response accuracy confidence index to perform node-by-node confidence analysis of the target viscosity curve and establish node confidence labels; configuring node lag influence factors using the node confidence labels; and performing adaptive feedback optimization of sequential nodes based on the node lag influence factors and difference comparison results.

[0041] Preferably, historical control response data of the printer is acquired, i.e., time-series data recorded during printer operation, which may include power control commands and actual response effects. This historical control response data is analyzed to quantify the response effects of historical control actions, and then a response accuracy confidence index is configured. If the response is highly consistent, with each increase in power viscosity resulting in a decrease, the response accuracy confidence index for power increases is high; conversely, if the response is highly inconsistent, the response accuracy confidence index is low. Then, the response accuracy confidence index is used to perform node-by-node confidence analysis of the target viscosity curve, i.e., analyzing the target viscosity value at each node of the target viscosity curve and determining the type and magnitude of the response control action. Node confidence labels are then established based on the configured response accuracy confidence index; if the historical confidence index of the required control action is high, the node is assigned a high confidence label, and if the historical confidence index of the required control action is low, the node is assigned a low confidence label. Next, node lag influence factors are configured using the node confidence labels to amplify or reduce the focus on future nodes. A larger node lag influence factor is configured for low-confidence nodes, and vice versa for high-confidence nodes. Finally, based on the node lag impact factor and the difference comparison results, the adaptive feedback optimization of the sequential nodes is performed, that is, to minimize the sum of the prediction errors of all nodes in the future prediction time domain. Here, the node lag impact factor gives higher weight to the prediction error of nodes marked as high lag risk, indicating that the optimizer prioritizes to ensure the future performance of high-risk nodes. Finally, the optimal control instruction sequence is output as the adaptive feedback optimization result.

[0042] Furthermore, the specific configuration of the feedback module 40 also includes configuring the lag cumulative impact term for each node based on the node lag impact factor and the difference comparison result; establishing a smoothing cost term between nodes; establishing a node objective function based on the lag cumulative impact term, the smoothing cost term, and the control energy cost term; and using the node objective function to perform adaptive feedback optimization management for each node.

[0043] Preferably, the lag cumulative impact term for each node is configured based on the node lag impact factor and the difference comparison results. The lag cumulative impact term penalizes future prediction errors. Specifically, the node lag impact factor is used as a weighting coefficient, with the weights configured according to the node confidence labels. The difference comparison results for each node are weighted, and the sum of squares of prediction errors for future nodes over the entire time domain is calculated as the lag cumulative impact term. The smoothing cost term penalizes drastic changes in control or state variables to ensure control stability and prevent overly aggressive actuator actions that could cause oscillations or overshoot. This is calculated as the sum of squares of the differences in control variables between adjacent nodes. The control energy cost term penalizes excessively large control variables themselves to ensure the control objective is achieved with minimal energy consumption, achieving high efficiency and energy saving. This is calculated as the sum of squares of the control variables for each node. Then, the lag cumulative impact term, smoothing cost term, and control energy cost term are weighted to determine the node objective function. The weights of these terms are used to balance tracking accuracy, control smoothness, and energy consumption.

[0044] Preferably, adaptive feedback optimization management is performed node by node using the node objective function. Specifically, in each control cycle, the current fused viscosity state and target trajectory are received, the future output is predicted based on different candidate control sequences, and the objective function value corresponding to each candidate control sequence is calculated using the node objective function. The optimizer optimizes and determines the candidate control sequence that minimizes the objective function value as the optimal control sequence, and outputs the first control quantity contained in the optimal control sequence to the actuator. The adaptive feedback optimization management is repeated in each control cycle, thereby achieving high-performance and high-efficiency tracking of the target viscosity curve.

[0045] The management module 50 is used to perform intelligent control management based on the adaptive feedback optimization results, which include light source power correction values, nozzle heating adjustment values, and pulse timing correction values.

[0046] Preferably, the management module is a logic control unit used to receive adaptive feedback optimization results, perform rationality verification and signal conversion, and generate precise control signals to achieve intelligent control and management of ink viscosity. The adaptive feedback optimization results include light source power correction values, nozzle heating adjustment values, and pulse timing correction values. The light source power correction value indicates the amount of power that needs to be increased or decreased based on the current power; the nozzle heating adjustment value indicates the amount of temperature that needs to be increased or decreased based on the current temperature setting; and the pulse timing correction value may include adjustments to the pulse width, interval, or frequency. Specifically, the light source power correction value, nozzle heating adjustment value, and pulse timing correction value are sent to the light source driver, respectively. The heater temperature controller and light source pulse controller management module ensure coordinated scheduling of multiple actuators. For example, just before the curing process begins, the heating temperature is slightly reduced to compensate for the viscosity surge caused by a sudden increase in light power, resulting in a smoother change in ink viscosity. Before the adjustment, the module checks whether the corrected light source power and temperature exceed the limits, ensuring that all control commands are executed within the safe range allowed by the hardware. After the adjustment, the module reads the actual status feedback of the actuators, such as the actual output power of the light source and the actual temperature of the heater, to ensure that the commands are executed correctly. If the execution effect does not meet expectations, an abnormal handling process is triggered to ensure that the printer is adjusted to the optimal working state, thereby improving print quality and process stability.

[0047] Furthermore, the variable viscosity UV ink system also includes a verification module, used to establish verification nodes based on the adaptive feedback optimization results, and to perform control verification based on the target viscosity curve using the verification nodes to establish verification feedback; and a feedback module, used to establish sequence compensation based on the verification feedback, and to perform node-by-node feedback optimization of the adaptive feedback optimization results using the sequence compensation.

[0048] Preferably, based on the adaptive feedback optimization results, specific nodes are selected as verification nodes in the printing path, such as periodic selection, selection of key process sections, or establishment of verification nodes after low-confidence nodes. The adaptive feedback optimization results and actual fusion viscosity values ​​corresponding to each verification node are recorded. At the verification node, the actual fusion viscosity value is compared and verified with the target viscosity value obtained from the target viscosity curve to obtain verification feedback, including the magnitude and direction of the verification error, control parameters such as printing speed, optical power, and temperature when the error occurs, and the corresponding verification confidence level obtained based on the measurement uncertainty of the sensor at that point. Then, the verification feedback is analyzed to establish sequence compensation, which is used to correct the offset sequence of future control commands. For example, if multiple consecutive verification nodes show consistent deviations, it is determined that there are systematic deviations such as model parameter drift and sensor calibration offset, and then the compensation amount is calculated in reverse. Finally, the sequence compensation is used to perform node-by-node feedback optimization of the adaptive feedback optimization results, that is, the nodes are optimized based on the sequence step size before adaptive feedback optimization, thereby eliminating recurring error patterns in advance and optimizing the overall control accuracy.

[0049] Furthermore, the variable viscosity UV ink system also includes an early warning module, used to identify deviations from the verification feedback, establish deviations from the threshold triggering identification result, and manage control anomaly reporting based on the deviations from the threshold triggering identification result.

[0050] Preferably, the early warning module continuously monitors the verification feedback signal and determines whether it is deviating from the normal state or is about to do so. Upon detecting an anomaly, it generates alarm information and executes preset management actions. Specifically, based on preset thresholds and rules, it identifies deviation threshold triggers in the verification feedback, including absolute value thresholds, statistical trend thresholds, and rate thresholds. This determines the deviation threshold trigger identification result, which may include alarm level, type, alarm information trigger source, and trigger-related parameter values. Finally, based on the deviation threshold trigger identification result, it manages the abnormal reporting, i.e., executing a preset tiered response strategy. For example, for low-level warnings, a Level 1 alert is issued, displaying a yellow warning message on the human-machine interface; for medium-level errors, a Level 2 intervention is issued, displaying a red alarm message and automatically pausing the printing process, switching to a safe degraded mode; for serious faults, a Level 3 abort is issued, immediately stopping the printing task and locking the system to prevent equipment damage or the production of large quantities of waste, while simultaneously notifying maintenance personnel via sound, light, and SMS.

[0051] In the above text, refer to Figure 1 A variable viscosity UV ink system according to embodiments of the present invention has been described in detail. Next, reference will be made to... Figure 2 A smart control method for variable viscosity UV ink according to embodiments of the present invention is described. The smart control method for variable viscosity UV ink, such as... Figure 2As shown, the method includes: performing multi-dimensional sensing data acquisition, wherein the multi-dimensional sensing data includes ink temperature, flow rate, surface tension, and photocuring process status; establishing an ink state feature vector based on the multi-dimensional sensing data; inputting the ink state feature vector into a dual-domain collaborative modeling unit containing an optical domain model and a rheological domain model, and outputting predicted viscosity distribution parameters; synchronously acquiring local viscosity data, performing authentication fusion based on the local viscosity data and predicted viscosity distribution parameters, and outputting fused viscosity state parameters; inputting the fused viscosity state parameters into a feedforward control layer, performing adaptive feedback optimization based on the difference comparison result between the target viscosity curve and the fused viscosity state parameters, and establishing an adaptive feedback optimization result; and performing intelligent control management based on the adaptive feedback optimization result, wherein the adaptive feedback optimization result includes light source power correction value, nozzle heating adjustment value, and pulse timing correction value.

[0052] In one possible implementation, inputting the fusion viscosity state parameter to the feedforward control layer includes: reading the printer's printing target data; performing target deposition path feature analysis based on the printing target data, wherein the target deposition path features include spraying speed features, linewidth features, and layer thickness requirement features; sending the target deposition path features to the feedforward control layer; using the fitting channel of the feedforward control layer to perform viscosity requirement fitting to establish a target viscosity curve; using the target viscosity curve and the fusion viscosity state parameter to perform node-by-node difference analysis to establish difference comparison results; and using the difference comparison results for adaptive feedback optimization.

[0053] In one possible implementation, the adaptive feedback optimization using the difference comparison results includes: acquiring historical control response data of the printer; configuring a response accuracy confidence index based on the historical control response data; performing node-by-node confidence analysis of the target viscosity curve using the response accuracy confidence index to establish node confidence labels; configuring node lag influence factors using the node confidence labels; and performing adaptive feedback optimization of sequential nodes based on the node lag influence factors and the difference comparison results.

[0054] In one possible implementation, the adaptive feedback optimization of sequential nodes based on the node lag impact factor and the difference comparison result includes: configuring a lag cumulative impact term for each node according to the node lag impact factor and the difference comparison result; establishing a smoothing cost term between nodes; establishing a node objective function based on the lag cumulative impact term, the smoothing cost term, and the control energy cost term; and using the node objective function to perform adaptive feedback optimization management for each node.

[0055] In one possible implementation, the step of inputting the ink state feature vector into a dual-domain collaborative modeling unit comprising an optical domain model and a rheological domain model, and outputting predicted viscosity distribution parameters, includes: parsing the ink state feature vector to obtain light source power, wavelength, pulse mode feature components, and ink temperature, flow rate, surface tension, and component concentration; inputting the light source power, wavelength, and pulse mode feature components into the optical domain model to output the photoinitiator activation rate; and after evaluating the molecular chain entanglement based on the ink temperature, flow rate, surface tension, and component concentration, inputting the molecular chain entanglement and photoinitiator activation rate into the rheological domain model to output predicted viscosity distribution parameters.

[0056] In one possible implementation, the authentication fusion based on the local viscosity data and the predicted viscosity distribution parameters includes: obtaining the hysteresis confidence of the local viscosity data; obtaining the prediction uncertainty of the predicted viscosity distribution parameters; and performing weighted fusion of the local viscosity data and the predicted viscosity distribution parameters based on the hysteresis confidence and the prediction uncertainty to establish fused viscosity state parameters.

[0057] In one possible implementation, the intelligent control method for the variable viscosity UV ink further includes: a verification module, used to establish verification nodes based on the adaptive feedback optimization results, and to perform control verification based on the target viscosity curve using the verification nodes to establish verification feedback; and a feedback module, used to establish sequence compensation based on the verification feedback, and to perform node-by-node feedback optimization of the adaptive feedback optimization results using the sequence compensation.

[0058] In one possible implementation, the intelligent control method for the variable viscosity UV ink further includes: an early warning module, used to identify deviation threshold triggers in the verification feedback, establish deviation threshold trigger identification results, and manage control anomaly reporting based on the deviation threshold trigger identification results.

[0059] In one possible implementation, the intelligent control method for the variable viscosity UV ink further includes: the fusion module includes a multi-point local viscosity sensor acquisition unit, which is used to perform weighted authentication on multiple local viscosity data in the same area, and output the redundant weighted authentication results as local viscosity data.

[0060] The variable viscosity UV ink system provided in this embodiment of the invention can execute the intelligent control method for variable viscosity UV ink provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0061] Although this application makes various references to certain modules in the system according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not used to limit the scope of protection of this invention.

[0062] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A variable viscosity UV ink system, characterized in that, The system includes: A multi-dimensional sensing module is used to perform multi-dimensional sensing data acquisition, including ink temperature, flow rate, surface tension, and photocuring process status. An ink state feature vector is established based on the multi-dimensional sensing data. The prediction module is used to input the ink state feature vector into a dual-domain collaborative modeling unit containing an optical domain model and a rheological domain model, and output predicted viscosity distribution parameters. The fusion module is used to synchronously collect local viscosity data, perform authentication fusion based on the local viscosity data and predicted viscosity distribution parameters, and output fused viscosity state parameters. The feedback module is used to input the fusion viscosity state parameters to the feedforward control layer, and to perform adaptive feedback optimization based on the difference comparison results between the target viscosity curve and the fusion viscosity state parameters, and to establish the adaptive feedback optimization result. The management module is used to perform intelligent control management based on the adaptive feedback optimization results, which include light source power correction values, nozzle heating adjustment values, and pulse timing correction values.

2. The variable viscosity UV ink system as described in claim 1, characterized in that, In the feedback module, the fusion viscosity state parameter is input to the feedforward control layer, including: Read the printer's printing target data, and analyze the target deposition path features based on the printing target data. The target deposition path features include spraying speed features, line width features, and layer thickness requirement features. The target deposition path features are sent to the feedforward control layer, and the viscosity requirement is fitted using the fitting channel of the feedforward control layer to establish the target viscosity curve. The target viscosity curve and the fusion viscosity state parameters are used to perform node-by-node difference analysis, establish difference comparison results, and use the difference comparison results for adaptive feedback optimization.

3. The variable viscosity UV ink system as described in claim 2, characterized in that, Adaptive feedback optimization using the difference comparison results includes: Acquire historical control response data of the printer, and configure the response accuracy confidence index based on the historical control response data; The confidence index of the response is used to perform node-by-node confidence analysis of the target viscosity curve and establish node confidence labels; Configure node hysteresis impact factors using the node confidence labels; Based on the node lag influence factor and the difference comparison results, an adaptive feedback optimization of the execution order nodes is performed.

4. The variable viscosity UV ink system as described in claim 3, characterized in that, Based on the node lag impact factor and the difference comparison results, an adaptive feedback optimization of the sequential nodes is performed, including: Configure the lag cumulative impact term for each node based on the node lag impact factor and the difference comparison result; Establish a smoothing cost term between nodes, and establish a node objective function based on the hysteresis cumulative effect term, smoothing cost term, and control energy cost term; The node objective function is used to perform adaptive feedback optimization management for each node.

5. The variable viscosity UV ink system as described in claim 1, characterized in that, In the prediction module, the ink state feature vector is input to a dual-domain collaborative modeling unit containing both an optical domain model and a rheological domain model, and the predicted viscosity distribution parameters are output, including: The ink state feature vector is analyzed to obtain the light source power, wavelength, pulse mode feature components, ink temperature, flow rate, surface tension, and component concentration. The light source power, wavelength, and pulse mode characteristic components are input into the optical domain model, and the photoinitiator activation rate is output. After assessing the molecular chain entanglement based on the ink temperature, flow rate, surface tension, and component concentration, the molecular chain entanglement and photoinitiator activation rate are input into the rheological domain model to output predicted viscosity distribution parameters.

6. The variable viscosity UV ink system as described in claim 1, characterized in that, The fusion module performs authentication fusion based on the local viscosity data and predicted viscosity distribution parameters, including: Acquire hysteresis confidence level for local viscosity data; Obtain the prediction uncertainty of the predicted viscosity distribution parameters, and perform weighted fusion of local viscosity data and predicted viscosity distribution parameters based on the hysteresis confidence and the prediction uncertainty to establish fused viscosity state parameters.

7. The variable viscosity UV ink system as described in claim 1, characterized in that, The system also includes: The verification module is used to establish verification nodes based on the adaptive feedback optimization results, use the verification nodes to perform control verification based on the target viscosity curve, and establish verification feedback. The feedback module is used to establish sequence compensation based on the verification feedback, and to perform node-by-node feedback optimization of the adaptive feedback optimization results using the sequence compensation.

8. The variable viscosity UV ink system as described in claim 7, characterized in that, The system also includes: The early warning module is used to identify deviations from the threshold trigger in the verification feedback, establish deviation threshold trigger identification results, and manage control anomaly reporting based on the deviation threshold trigger identification results.

9. The variable viscosity UV ink system as claimed in claim 1, characterized in that, The fusion module includes a multi-point local viscosity sensor acquisition unit, which is used to perform weighted authentication on multiple local viscosity data in the same area and output the redundant weighted authentication results as local viscosity data.

10. A smart control method for variable viscosity UV ink, characterized in that, The method is applied to the variable viscosity UV ink system according to any one of claims 1-9, and the method includes: Perform multi-dimensional sensing data acquisition, the multi-dimensional sensing data including ink temperature, flow rate, surface tension, and photocuring process status, and establish an ink state feature vector based on the multi-dimensional sensing data; The ink state feature vector is input into a dual-domain collaborative modeling unit containing both an optical domain model and a rheological domain model, and the predicted viscosity distribution parameters are output. Local viscosity data is collected synchronously, and the local viscosity data and predicted viscosity distribution parameters are used for authentication fusion to output fused viscosity state parameters. The fusion viscosity state parameters are input to the feedforward control layer. Based on the difference comparison between the target viscosity curve and the fusion viscosity state parameters, adaptive feedback optimization is performed to establish the adaptive feedback optimization result. Intelligent control management is performed based on the adaptive feedback optimization results, which include light source power correction values, nozzle heating adjustment values, and pulse timing correction values.

Images