Circulating ink supply intelligent control method and system based on UV printer

Abstract

The invention discloses a circular ink supply intelligent control method and system based on a UV printer, and the method comprises the steps: S1, collecting a digital image file of a to-be-printed task, and obtaining the consumption prediction data of ink of each color through the analysis of pixel information; s2, based on the consumption prediction data, obtaining a flow demand through a Poiseuille law fluid dynamic conversion algorithm, and based on the flow demand, constructing an ink supply control regression model to carry out ink supply coupling feed-forward calculation to obtain a feed-forward control parameter set; s3, constructing a self-learning database for input comparison, and adjusting and optimizing the feedforward control parameter set based on unscented Kalman filtering according to the historical control effect deviation to generate a final control instruction; and S4, intelligent control over circulating ink supply is conducted on an execution mechanism of the circulating ink supply system based on the final control instruction, the color rendition degree and consistency of a printed picture are improved, and the operation cost is reduced while the ink supply quality is guaranteed.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology, and in particular to an intelligent control method and system for circulating ink supply based on a UV printer. Background Technology

[0002] As an important industrial printing device, UV printers have evolved their ink supply systems from simple open-loop control to preliminary closed-loop regulation. Early systems often used mechanical pumps and valves with fixed parameters, relying on manual experience to set the ink supply volume, lacking real-time responsiveness to the printing task. With increased digitalization, some systems began to introduce ink consumption prediction based on image pixel information and attempted to use fluid dynamics principles (such as Poiseuille's law) for preliminary estimation of flow requirements. In recent years, to further improve control accuracy, some solutions have begun to introduce feedback mechanisms based on historical data, using simple PID controllers or lookup tables to fine-tune ink supply parameters to adapt to changes in ink viscosity, ambient temperature, and other operating conditions.

[0003] Currently, although the circulating ink supply system has achieved initial intelligence, it still faces the following problems in practical applications: Insufficient accuracy in nonlinear characteristic modeling: There is strong nonlinear coupling between core parameters of the ink supply system such as pressure, flow rate and viscosity. The ink viscosity changes exponentially with the ambient temperature. The relationship between flow rate and pressure is affected by pipeline resistance and ink pump characteristics, exhibiting piecewise nonlinearity. Traditional linear models or simple nonlinear fitting such as polynomial regression cannot accurately characterize these complex relationships, resulting in large deviations in feedforward parameter prediction. Weak dynamic scene response capability: During printing tasks, the flow demand may change abruptly by 2-3 times due to the complexity of the pattern, such as switching from text to large-format color blocks. However, the control parameter adjustment of the existing system relies on fixed thresholds or delayed feedback, which can easily lead to ink supply lag or overshoot. For example, when printing large-format black blocks, the flow demand suddenly increases, but the ink pump speed adjustment is delayed by more than 0.5 seconds, resulting in insufficient ink supply in the initial stage and the color blocks appearing white. Therefore, this paper proposes an intelligent control method and system for circulating ink supply based on UV printers to solve the above problems. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art and to achieve the above objectives, the present invention proposes the following technical solution: The intelligent control method for circulating ink supply based on UV printers includes: S1: Acquire digital image files of the task to be printed, and obtain predicted data on the consumption of each color ink by analyzing pixel information; S2: Based on the consumption prediction data, the flow demand is obtained through the Poiseuille law fluid dynamics conversion algorithm. Based on the flow demand, an ink supply control regression model is constructed to perform ink supply coupling feedforward calculation and obtain the set of feedforward control parameters. S3: Construct a self-learning database for input comparison, and adjust and optimize the feedforward control parameter set based on unscented Kalman filtering according to the historical control effect deviation to generate the final control command; S4: Perform intelligent control of the circulating ink supply system's actuators based on the final control command.

[0005] The process of obtaining consumption forecast data is as follows: The process involves acquiring a digital image file of the print job, analyzing the RGB color information of its pixels, converting RGB to CMYK components using a color space conversion formula, calculating the ink consumption of each color in the entire image, and summing the ink components of all pixels to obtain predicted ink consumption data. , represented as: ; in, It is the width of the image. For high pixel count, Coordinates The value of the i-th ink component of the pixel. This represents the predicted consumption data for the i-th type of ink.

[0006] The process of acquiring traffic demand is as follows: Based on Poiseuille's law and a fluid dynamics conversion algorithm, the ink consumption prediction data was then used. Converting this to the ink supply system's flow requirements, let the printer's printing speed be v and the print width be W. Then, what are the required ink flow rate per unit time and the predicted ink consumption data? The relational formula is expressed as: ; in, Let i be the flow rate requirement for the i-th type of ink. It is the width of the image. For high pixel count, Indicates printing speed.

[0007] The ink supply control regression model includes a pump speed control output section, a valve control output section, and a valve control output section. The pump speed control section outputs the pressure control parameters for the i-th type of ink. The pump speed control output section outputs the control parameters for the pump speed corresponding to the i-th type of ink. The valve control output section outputs the valve control parameters corresponding to the i-th type of ink. .

[0008] Pressure control parameters for the i-th type of ink The acquisition process is as follows: The pressure control parameters are obtained based on the nonlinear kernel function of SVR. The process is as follows: ; in, This represents the pressure control parameter for the i-th type of ink output by the model. This represents a nonlinear kernel function based on SVR. This indicates the dynamic viscosity of the ink.

[0009] The control parameters for the pump speed corresponding to the i-th type of ink The acquisition process is as follows: Based on traffic demand If the pump speed and flow rate are considered to have a linear relationship within the high-efficiency operating range, the calculation process for the pump speed control parameters is as follows: ; in, This is the proportionality coefficient for the pump speed corresponding to the i-th type of ink. This represents the intercept or offset of the pump speed corresponding to the i-th ink type. This represents the control parameter for the pump speed corresponding to the i-th type of ink.

[0010] Valve control parameters corresponding to the i-th type of ink The acquisition process is as follows: Record the traffic demand after the system traffic stabilizes at each opening degree α. The collected traffic demand Plot the data on a coordinate graph, with the opening degree α on the horizontal axis and the flow demand on the vertical axis. Fit these data points, dividing the entire interval into multiple segments, and fitting each segment with a quadratic function to obtain the relationship function. Based on relational functions Fitting the relationship between valve opening and flow rate The valve control parameters for obtaining the valve opening degree corresponding to the i-th type of ink are obtained. .

[0011] The process of adjusting and optimizing the set of feedforward control parameters is as follows: The feedforward control parameters to be optimized are defined as state vectors. Based on state vector Construct state equations and observation equations, define Sigma points, and substitute each Sigma point into the state equations and observation equations to obtain the predicted control parameters. Deviation from predicted control effect ; Based on control parameters Constructing the prior covariance matrix of ink supply Based on control effect deviation Constructing the covariance matrix of ink supply observations Based on the ink supply observation covariance matrix With ink supply prior covariance matrix The Kalman gain is obtained by the following formula: , Kalman gain; After each ink supply cycle is completed, the control effect deviation of the i-th ink supply task is... Compared with predicted observations The difference is calculated to obtain the deviation factor. The state is updated based on the Kalman gain and the measured deviation of the actual control effect in the new task, and is expressed as follows: ; in, This represents the state vector updated in the previous step. This represents the latest state vector, i.e., the final control command after optimization and update.

[0012] The intelligent control system for circulating ink supply based on UV printers includes: Status acquisition module: Acquires digital image files of the task to be printed, and obtains predicted data on the consumption of each color ink by analyzing pixel information; Control feedback module: Based on consumption prediction data, the flow demand is obtained through the fluid dynamics conversion algorithm of Poiseuille's law. Based on the flow demand, an ink supply control regression model is constructed to perform ink supply coupling feedforward calculation and obtain a set of feedforward control parameters. Command decision module: Constructs a self-learning database for input comparison, and uses unscented Kalman filtering to adjust and optimize the set of feedforward control parameters based on historical control effect deviations to generate the final control command; Ink supply control module: performs intelligent control of the cyclic ink supply system's actuators based on the final control commands.

[0013] The present invention has the following beneficial effects: In this invention, firstly, by modeling with the RBF kernel function based on SVR, the complex nonlinear relationship between ink viscosity, ambient temperature, flow rate demand and parameters such as pressure, rotation speed, and valve opening in the circulating ink supply system can be effectively characterized. This solves the problem of insufficient fitting of nonlinear characteristics by traditional models, making the prediction of feedforward control parameters more consistent with the actual ink supply law, and fundamentally improving the basic accuracy of parameter setting.

[0014] Secondly, by combining unscented Kalman filtering to optimize feedforward parameters in real time, it can quickly respond to dynamic scenarios such as sudden changes in flow rate, temperature fluctuations, and changes in ink characteristics during the printing process. By adjusting the control commands of the actuator in real time, it can avoid excessive or insufficient ink supply, and ensure that the ink supply status always matches the printing demand when the working conditions change, thus maintaining stable print quality. Finally, by distributing multi-dimensional instructions and coordinating the control of the execution mechanism, combined with precise modeling of the coupling relationship of multi-color parameters, the superposition and amplification of ink supply deviations of each color can be effectively avoided, ensuring the accurate color ratio of mixed inks, improving the color reproduction and consistency of the printed image, and reducing operating costs while ensuring ink supply quality. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating the steps of the intelligent control method and system for circulating ink supply based on a UV printer proposed in this invention.

[0016] Figure 2 This is a system block diagram of the intelligent control method and system for circulating ink supply based on a UV printer proposed in this invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Example 1 like Figure 1 As shown, the intelligent control method for circulating ink supply based on a UV printer proposed in this invention includes: S1: Acquire digital image files of the task to be printed, and obtain predicted data on the consumption of each color ink by analyzing pixel information; It receives digital files for printing tasks, such as PDF, TIFF, and dedicated RIP files, parses them into bitmap format, and extracts the original color information of each pixel, including RGB (red, green, and blue) color values. The system acquires digital image files of the task to be printed, analyzes the RGB color information of its pixels, converts RGB to CMYK components using color space conversion formulas, and calculates the consumption of each ink color in the entire image. The consumption prediction data is obtained by summing the ink color components of all pixels. , represented as: ; in, It is the width of the image. For high pixel count, Coordinates The value of the i-th ink component of the pixel. This represents the predicted consumption data for the i-th type of ink; Specifically, the summation yields... The theoretical total ink volume value, combined with the subsequent ink supply coupling feedforward model, is used to calculate the instantaneous consumption rate curve of each ink color over time during the printing process, providing important time series data for subsequent feedforward ink supply control.

[0019] S2: Based on the consumption prediction data, the flow demand is obtained by the fluid dynamics conversion algorithm based on Poiseuille's law. Based on the flow demand, the ink supply control regression model is combined to perform ink supply coupling feedforward calculation to obtain the set of feedforward control parameters. In an ink supply system, the flow of ink follows the laws of fluid dynamics. There is a clear physical relationship between the flow rate, pressure, and velocity of ink in the ink supply pipeline. For the pipeline of the ink supply system, when the ink is in a laminar flow state, it can be described by Poiseuille's law. The basic formula of the fluid dynamics conversion algorithm based on Poiseuille's law is expressed as follows: ; in, This indicates the volumetric flow rate of ink in the ink supply line. This represents the pressure difference between the two ends of the pipeline, and r represents the inner diameter of the ink supply pipeline. The value represents the dynamic viscosity of the ink, and L represents the length of the ink supply line. Furthermore, by establishing a physical bridge between the flow rate and various key parameters in the ink supply system, it was revealed how the flow rate of ink in the pipeline is affected by factors such as the pressure difference between the two ends of the pipeline, the inner diameter of the pipeline, the dynamic viscosity of the ink itself, and the length of the pipeline when the ink flows in the pipeline under laminar flow conditions. This provides the most basic theoretical basis for subsequent flow control. Specifically, the fluid dynamics conversion algorithm based on Poiseuille's law is grounded in the ink consumption prediction data obtained in step S1. This total ink volume data is static and cannot be directly used for real-time control, so it needs to be... Converted into instantaneous traffic demand; Then, the ink consumption prediction data Converting this to the ink supply system's flow requirements, let the printer's printing speed be v and the print width be W. Then, what are the required ink flow rate per unit time and the predicted ink consumption data? The relational formula is expressed as: ; in, Let i be the flow rate requirement for the i-th type of ink. It is the width of the image. For high pixel count, Indicates printing speed; Specifically, ink consumption forecasting data enables the transformation from static to dynamic real-time flow demand. It calculates the total ink consumption for the entire printing task, distributing it to each second to obtain the flow requirement per unit time. This allows the ink supply system to adjust the ink supply in real time according to the printing rhythm, ensuring that the ink supply is synchronized with the printing process. The process of obtaining the ink supply control regression model is as follows: Under different ink viscosities (μ) and ambient temperatures (T), a series of known flow rate requirements are applied. A training dataset is created, and an ink supply control regression model is built based on this dataset, including pressure control output, pump speed control output, and valve control output. The pressure control output is represented as follows: The pressure control output section outputs pressure control parameters corresponding to different inks, implemented based on the nonlinear kernel function of SVR. The process is as follows: ; in, This represents the pressure control parameter for the i-th type of ink output by the model. This represents a nonlinear kernel function based on SVR. Indicates the dynamic viscosity of the ink; Specifically, the output of the ink supply control regression model It is used to accurately fit the complex relationship between flow rate demand and pressure control parameters, while incorporating ink viscosity. The relationship between pressure and flow rate in actual ink supply systems is not a simple linear one, but is also affected by factors such as ink viscosity changes with temperature. Traditional linear models struggle to accurately describe this relationship, while ink supply control regression models, based on known flow rate requirements, address the issue. Data training can capture these complex nonlinear relationships and directly output the most suitable pressure control parameters based on the current flow requirements, ink viscosity and ambient temperature, so that the pressure adjustment can adapt to various complex working conditions and ensure that the flow rate can remain stable near the required value under different conditions. The pump speed control output is expressed as follows: Simultaneously based on traffic demand The ink supply control regression model treats the pump speed and flow rate as having a linear relationship within the high-efficiency operating range. Therefore, the calculation process for the pump speed control parameters is as follows: ; in, The proportionality coefficient for the pump speed corresponding to the i-th ink type represents the increase in flow rate that can be provided for each unit increase in pump speed. This represents the intercept or offset of the pump speed corresponding to the i-th ink type, and indicates internal leakage and mechanical friction within the pump. This represents the control parameter indicating the pump speed corresponding to the i-th type of ink; Specifically, the pump speed control output of the ink supply control regression model transforms the flow demand into specific speed control commands for the ink supply pump. As the power source for delivering ink, the pump's speed directly affects the output flow rate. Within the pump's efficient operating range, speed and flow rate exhibit a stable linear relationship, based on known flow demand. and the coefficients obtained from calibration , Calculate the required pump speed. This allows the pump to operate precisely according to flow requirements, providing the most suitable power for ink delivery; The valve control output is represented as follows: By fitting the relationship between valve opening and flow rate , The function represents the relational function, and its inverse function yields the valve control parameters for the valve opening degree corresponding to the i-th type of ink. ; Among them, valve control parameters The size of the flow channel cross-section determines the fluid flow capacity; a larger opening results in lower flow resistance and a higher flow rate under the same pressure difference. The larger the value, the more necessary it is to fit the characteristic curve relationship. This is achieved through a fitting method; The valve is changed from fully closed (α=0%) to fully open (α=100%), with the valve opening α changed in fixed steps (5%). At each opening, after the system flow stabilizes, the flow demand corresponding to the valve opening α is recorded. This yields data points covering the entire operating range of the valve. ; The collected data points were plotted on a coordinate graph, with the opening degree α on the horizontal axis and the flow demand on the vertical axis. Mathematical methods were used to fit these data points, dividing the entire interval into multiple segments, and fitting each segment with a quadratic function to obtain a function that reflects the distribution pattern of the data points. , This function represents the valve's flow characteristics; pass Obtain flow demand and valve opening The relationship can be used to deduce the corresponding valve control parameters. ; Finally, a corresponding subset of control instructions is generated for each ink color. All subsets of ink colors together constitute the set of feedforward control parameters.

[0020] S3: Construct a self-learning database for input comparison, and adjust and optimize the feedforward control parameter set based on unscented Kalman filtering according to the historical control effect deviation to generate the final control command; The self-learning data construction process is as follows: Continuously record the initial control commands generated by S2 during each print task. And the system status measured by sensors after the initial control commands are executed, including the corresponding actual flow rate. ; Create a database with the following record format: record t=( e) where, Let represent the set of feedforward control parameters, and e represent the control effect deviation, where represents the control effect deviation of the i-th ink. ,in, It is the actual consumption of the i-th type of ink. This is ink consumption forecast data; When a new printing task begins, the system generates an initial set of feedforward control parameters, queries similar task records in the self-learning database, and adjusts and optimizes the control parameters based on the deviation of historical control effects using the joint parameter estimation method of unscented Kalman filtering. The process of adjusting and optimizing the set of feedforward control parameters is as follows: The feedforward control parameters to be optimized are defined as state vectors. Each parameter in the feedforward control parameters corresponds to an execution dimension of the ink supply system; Based on state vector Construct the state equation and observation equation, and the state vector. Lacking active evolution characteristics, meaning the parameters themselves do not change spontaneously, process noise is first added. This reflects the uncertainties in the optimization process, such as differences in operating conditions between different printing tasks, the cumulative effect of small sensor errors, and process noise. It needs to be set to zero-mean Gaussian noise, and its covariance matrix needs to be calibrated according to the stability experiment of the ink supply system, such as the fluctuation range of parameters in multiple repeated printing tasks. The state equation is expressed as: ,in, This represents the state vector from the previous step. The observation equation is expressed as: ,in, Let be the observation function, representing the nonlinear mapping relationship between the feedforward control parameters and the deviation. Indicates observation noise. This indicates the control effect deviation of the i-th ink supply task; Then, the feedforward control parameters of the unscented Kalman filter are adjusted. Sigma points are defined based on the state vector, and each Sigma point is substituted into the state equation of the circulating ink supply system. In this process, the predicted control parameters are obtained. The state vector here directly corresponds to the set of feedforward control parameters for the cyclic ink supply; Specifically, Sigma points are a set of discrete sampling points used in unscented Kalman filtering to approximate the state probability distribution. These points are obtained by analyzing the state vector. The mean and covariance matrices are generated by unscented transformation, which can accurately capture the state distribution characteristics of nonlinear systems with a small number of points, providing a discretized computational basis for subsequent prediction and update steps. Predicted Sigma points The observation function is substituted into the observation equation of the circulating ink supply system. In the process, the predicted control effect deviation is obtained. This deviation reflects the difference between the actual consumption and the predicted consumption during the circulating ink supply process; For prediction By performing a weighted average, the prior state estimate of the ink supply system is obtained. and the corresponding ink supply prior covariance matrix ; Specifically, the prior covariance matrix of ink supply By generating each predicted Sigma point Compared with the final calculated prior state estimate of ink supply (i.e., the weighted average center of all points) is compared to obtain the deviation of each point from the center, and then based on each Sigma point By weighting these deviations using pre-defined weights and combining the results calculated for all Sigma points in the previous step, we obtain the ink supply prior covariance matrix. The diagonal elements of this matrix represent the uncertainty of each state component, such as pressure setting and pump speed, while the off-diagonal elements represent whether there is a correlation between the uncertainties of different state components. For example, does inaccurate pressure prediction lead to inaccurate pump speed prediction? Deviation of predicted control effect The points are weighted and averaged to obtain the predicted ink supply observations. and ink supply observation covariance matrix These statistics provide a basis for subsequent optimization of the circulating ink supply parameters; Specifically, the covariance matrix of the ink supply observations It represents a measure of uncertainty regarding the observed predicted value (i.e., the deviation predicted value), expressed as a Sigma point for each predicted deviation. (Representing a possible observed bias) is compared with the final calculated predicted observation (i.e., the center of the weighted average of all bias points) to obtain the bias of each bias point relative to the center. Then, according to the weights, the weighted sum of the squares of these observed biases is calculated. Combining these results yields the observation covariance matrix. ; Based on ink supply observation covariance matrix With ink supply prior covariance matrix The Kalman gain is obtained by the following formula: ; After each ink supply cycle is completed, the control effect deviation of the i-th ink supply task is... Compared with predicted observations The difference is calculated to obtain the deviation factor. , (i.e., the difference between the actual consumption and the predicted consumption of each ink color), deviation factor The magnitude of the error directly reflects the degree of prediction deviation of the feedforward parameters in the circulating ink supply system; Based on the Kalman gain, the state is updated by combining the measured deviation of the actual control effect of the new task, and expressed as follows: ; in, This represents the final control command, which is the optimized and updated set of feedforward control commands. Specifically, the final control command is obtained through trace Kalman filtering, which integrates the latest and optimal set of feedforward control parameters that incorporate historical control performance deviations, including a subset of optimized control commands. Each value in the vector is the target instruction that the control system needs to send directly to the corresponding actuator (such as a pressure regulator or pump driver) when executing this printing task.

[0021] S4: Intelligent control of the circulating ink supply system's actuators based on final control commands; Based on the final control command, it is broken down by ink color and actuator type, including pressure parameters sent to the electronic pressure valve of the corresponding ink color, speed parameters sent to the servo ink pump of each ink color, and opening parameters sent to the proportional solenoid valve of the corresponding pipeline. The coordinated action and dynamic adaptation of the actuators are started synchronously according to the instructions. The servo ink pump draws ink from the ink cartridge at the target speed. The electronic pressure valve adjusts the pipeline pressure to the target value. The proportional solenoid valve controls the flow rate of ink into the mixing chamber by the opening degree. For example, in the final control commands, taking black ink as an example, the final control commands for black ink are as follows: , , The pressure valve, servo pump, and solenoid valve of the black pipeline are driven respectively to perform intelligent control of the circulating ink supply.

[0022] Example 2 like Figure 2 As shown, the intelligent control system for circulating ink supply based on a UV printer includes: Status acquisition module: Acquires digital image files of the task to be printed, and obtains predicted data on the consumption of each color ink by analyzing pixel information; Control feedback module: Based on consumption prediction data, the flow demand is obtained through the fluid dynamics conversion algorithm of Poiseuille's law. Based on the flow demand, an ink supply control regression model is constructed to perform ink supply coupling feedforward calculation and obtain a set of feedforward control parameters. Command decision module: Constructs a self-learning database for input comparison, and uses unscented Kalman filtering to adjust and optimize the set of feedforward control parameters based on historical control effect deviations to generate the final control command; Ink supply control module: performs intelligent control of the cyclic ink supply system's actuators based on the final control commands.

[0023] In the application, several formulas are calculated by removing dimensions and taking their numerical values. The formulas are established by collecting a large amount of data and simulating the most recent real situation. Some coefficients or weights in the formulas are set by those skilled in the art according to the actual situation, so they will not be elaborated here.

[0024] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0025] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for intelligent control of circulating ink supply based on a UV printer, characterized in that, include: S1: Acquire digital image files of the task to be printed, and obtain predicted data on the consumption of each color ink by analyzing pixel information; S2: Based on the consumption prediction data, the flow demand is obtained through the Poiseuille law fluid dynamics conversion algorithm. Based on the flow demand, an ink supply control regression model is constructed to perform ink supply coupling feedforward calculation and obtain the set of feedforward control parameters. S3: Construct a self-learning database for input comparison, and adjust and optimize the feedforward control parameter set based on unscented Kalman filtering according to the historical control effect deviation to generate the final control command; S4: Perform intelligent control of the circulating ink supply system's actuators based on the final control command.

2. The intelligent control method for circulating ink supply based on a UV printer according to claim 1, characterized in that, The process of obtaining consumption forecast data is as follows: The process involves acquiring a digital image file of the print job, analyzing the RGB color information of its pixels, converting RGB to CMYK components using a color space conversion formula, calculating the ink consumption of each color in the entire image, and summing the ink components of all pixels to obtain predicted ink consumption data. , represented as: ； in, It is the width of the image. For high pixel count, Coordinates The value of the i-th ink color component of the pixel. This represents the predicted consumption data for the i-th type of ink.

3. The intelligent control method for circulating ink supply based on a UV printer according to claim 2, characterized in that, The process of acquiring traffic demand is as follows: Based on Poiseuille's law and a fluid dynamics conversion algorithm, the ink consumption prediction data was then used. Converting this to the ink supply system's flow requirements, and assuming the printer's printing speed is v, the required ink flow rate per unit time and the predicted ink consumption data are as follows: The relational formula is expressed as: ; in, Let i be the flow rate requirement for the i-th type of ink. It is the width of the image. For high pixel count, Indicates printing speed.

4. The intelligent control method for circulating ink supply based on a UV printer according to claim 1, characterized in that, The ink supply control regression model includes a pump speed control output section, a valve control output section, and a valve control output section. The pump speed control section outputs the pressure control parameters for the i-th type of ink. The pump speed control output section outputs the control parameters for the pump speed corresponding to the i-th type of ink. The valve control output section outputs the valve control parameters corresponding to the i-th type of ink. .

5. The intelligent control method for circulating ink supply based on a UV printer according to claim 4, characterized in that, Pressure control parameters for the i-th type of ink The acquisition process is as follows: The pressure control parameters are obtained based on the nonlinear kernel function of SVR. The process is as follows: ； in, This represents the pressure control parameter for the i-th type of ink output by the model. This represents a nonlinear kernel function based on SVR. This indicates the dynamic viscosity of the ink.

6. The intelligent control method for circulating ink supply based on a UV printer according to claim 4, characterized in that, The control parameters for the pump speed corresponding to the i-th type of ink The acquisition process is as follows: Based on traffic demand If the pump speed and flow rate are considered to have a linear relationship within the high-efficiency operating range, the calculation process for the pump speed control parameters is as follows: ; in, This is the proportionality coefficient for the pump speed corresponding to the i-th type of ink. This represents the intercept or offset of the pump speed corresponding to the i-th ink type. This represents the control parameter for the pump speed corresponding to the i-th type of ink.

7. The intelligent control method for circulating ink supply based on a UV printer according to claim 4, characterized in that, Valve control parameters corresponding to the i-th type of ink The acquisition process is as follows: Record the traffic demand after the system traffic stabilizes at each opening degree α. The collected traffic demand Plot the data on a coordinate graph, with the opening degree α on the horizontal axis and the flow demand on the vertical axis. Fit these data points, dividing the entire interval into multiple segments, and fitting each segment with a quadratic function to obtain the relationship function. Based on relational functions Fitting the relationship between valve opening and flow rate The valve control parameters for obtaining the valve opening degree corresponding to the i-th type of ink are obtained. .

8. The intelligent control method for circulating ink supply based on a UV printer according to claim 7, characterized in that, The process of adjusting and optimizing the set of feedforward control parameters is as follows: The feedforward control parameters to be optimized are defined as state vectors. Based on state vector Construct state equations and observation equations, define Sigma points, and substitute each Sigma point into the state equations and observation equations to obtain the predicted control parameters. Deviation from predicted control effect ; Based on control parameters Constructing the prior covariance matrix of ink supply Based on control effect deviation Constructing the covariance matrix of ink supply observations Based on the ink supply observation covariance matrix With ink supply prior covariance matrix The Kalman gain is obtained by the following formula: , Kalman gain; After each ink supply cycle is completed, the control effect deviation of the i-th ink supply task is... Compared with predicted observations The difference is calculated to obtain the deviation factor. The state is updated based on the Kalman gain and the measured deviation of the actual control effect in the new task, and is expressed as follows: ; in, This represents the state vector updated in the previous step. This represents the latest state vector, i.e., the final control command after optimization and update.

9. A circulating ink supply intelligent control system based on a UV printer, implemented according to any one of claims 1-8, characterized in that, include: Status acquisition module: Acquires digital image files of the task to be printed, and obtains predicted data on the consumption of each color ink by analyzing pixel information; Control feedback module: Based on consumption prediction data, the flow demand is obtained through the fluid dynamics conversion algorithm of Poiseuille's law. Based on the flow demand, an ink supply control regression model is constructed to perform ink supply coupling feedforward calculation and obtain a set of feedforward control parameters. Command decision module: Constructs a self-learning database for input comparison, and uses unscented Kalman filtering to adjust and optimize the set of feedforward control parameters based on historical control effect deviations to generate the final control command; Ink supply control module: performs intelligent control of the cyclic ink supply system's actuators based on the final control commands.

Images