Ink supply control method, system, device and storage medium
By constructing a dynamic model of the ink supply system and combining feedback and feedforward control quantities, the problems of high sensor cost, poor compatibility and low accuracy in existing ink supply systems are solved, and a high-precision and stable ink supply effect is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-26
- Publication Date
- 2026-03-10
AI Technical Summary
Existing ink supply systems suffer from problems such as high cost, poor sensor compatibility, easy damage, low accuracy, and poor anti-interference in flow and pressure control, which are particularly prominent in multi-channel control systems.
By constructing a dynamic model of the ink supply system, the measurement status of the ink cartridge and ink pump is obtained in real time. Combined with data from liquid level, air pressure and speed sensors, the ink pump control signal is calculated using feedback and feedforward control quantities to achieve compensation for load changes and fluid characteristic fluctuations.
It improves the control accuracy and anti-interference capability of the ink supply system, achieves high-precision and stable ink supply, and reduces problems such as ink interruption and uneven spraying.
Smart Images

Figure CN121375332B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of fluid control technology, and in particular to an ink supply control method, system, device and storage medium. Background Technology
[0002] The ink supply system is a crucial subsystem of inkjet printing equipment. Based on the ink piping path, ink supply systems can be categorized into circulating and non-circulating systems. Ensuring stable flow or pressure between the ink cartridge and the printhead is a critical prerequisite for stable printing. Conventional methods maintain stable ink cartridge pressure or adjust the speed of the ink pump in the ink supply branch to ensure stable flow at the printhead's ink inlet. Both methods require measuring the pressure or flow at the printhead's ink inlet as feedback information to achieve closed-loop control, but they have the following limitations: flow sensors are expensive, especially high-precision models, significantly increasing equipment costs, a problem particularly pronounced in multi-channel control systems; contact-type flow / pressure sensors face material compatibility challenges when handling corrosive or high-viscosity fluids, and are easily damaged or contaminated; to measure ink supply pressure and flow or add an ink pump, an additional node needs to be added upstream of the printhead, increasing the risk of leakage and pressure instability; and using the method of adjusting the speed of the ink pump in the ink supply branch to ensure flow can easily introduce cavitation effects and tiny air bubbles into the piping, affecting print quality.
[0003] To avoid the aforementioned problems, some existing technologies employ open-loop pump speed control, which indirectly controls the flow rate by precisely controlling the pump speed. This method has the lowest cost, but its accuracy depends entirely on the pump's volumetric efficiency and has no ability to compensate for load changes, fluctuations in fluid characteristics (such as viscosity and temperature), and pump wear. It also has poor control accuracy and anti-interference capabilities. Some existing technologies also employ static characteristic lookup table schemes, which estimate the flow rate using a pre-calibrated pump pressure-flow-speed mapping table. Its accuracy is heavily dependent on the completeness and invariance of the calibration conditions and cannot adapt to changes in fluid characteristics and long-term degradation of pump performance, lacking self-adaptive capabilities. Summary of the Invention
[0004] The present invention aims to improve at least one technical problem in the prior art.
[0005] The first aspect of the present invention provides an ink supply control method, applied to an ink supply control system, the ink supply control system comprising: a control device, an ink supply cartridge, and an ink pump, wherein the outlet of the ink pump is connected to the inlet of the ink supply cartridge, and the control device is electrically connected to the ink pump.
[0006] The ink supply control method includes the following steps:
[0007] Obtain the current ink cartridge measurement status and the current actual operating speed of the ink pump;
[0008] Obtain historical ink cartridge measurement status and historical ink pump actual operating speed;
[0009] The preset dynamic model of the ink supply system is updated based on the current ink cartridge measurement status, the current actual ink pump operating speed, the historical ink cartridge measurement status, and the historical actual ink pump operating speed.
[0010] The feedback control quantity is calculated based on the current ink supply cartridge measurement status, the preset ink supply cartridge target status, and the preset feedback coefficient.
[0011] Based on the updated dynamic model of the ink supply system, the current measurement status of the ink cartridge, the current actual operating speed of the ink pump, the historical measurement status of the ink cartridge, and the historical actual operating speed of the ink pump, the feedforward control quantity is calculated.
[0012] The ink pump control signal is calculated based on the feedback control quantity and the feedforward control quantity.
[0013] Optionally, in a first implementation of the first aspect of the present invention, the ink supply control system further includes a liquid level sensor, a pressure sensor, and a speed feedback sensor. The liquid level sensor is used to acquire liquid level distribution information at at least two liquid level measurement points of the ink supply cartridge. The pressure sensor is used to acquire the liquid surface pressure of the ink supply cartridge. The speed feedback sensor is used to acquire the real-time speed of the ink pump. The control device is electrically connected to the liquid level sensor, the pressure sensor, and the speed feedback sensor. Acquiring the current ink supply cartridge measurement status and the current actual operating speed of the ink pump includes:
[0014] Control the liquid level sensor to obtain the current liquid level at each liquid level measurement point;
[0015] Control the air pressure sensor to obtain the current liquid level air pressure;
[0016] The speed feedback sensor is used to obtain the current actual operating speed of the ink pump.
[0017] Based on the current liquid level at each liquid level measurement point, the current discrete liquid level distribution information is obtained;
[0018] Based on the current discrete liquid level distribution information and the discrete liquid level distribution information at the previous moment, the current liquid level state transition vector is obtained;
[0019] The current ink cartridge measurement status is generated based on the current liquid level pressure and the current liquid level state transfer vector.
[0020] Optionally, in a second implementation of the first aspect of the present invention, obtaining the current discrete liquid level distribution information based on the current liquid level at each liquid level measurement point includes:
[0021] For each liquid level measurement point, if the current liquid level is not lower than the measurement position, the current liquid level distribution status is marked as true;
[0022] If the current liquid level is lower than the measurement position, the current liquid level distribution status is marked as false;
[0023] Based on the current liquid level distribution status of each liquid level measurement point, the current discrete liquid level distribution information is generated.
[0024] Optionally, in a third implementation of the first aspect of the present invention, obtaining the current liquid level state transition vector based on the current discrete liquid level distribution information and the discrete liquid level distribution information at the previous moment includes:
[0025] Based on the current discrete liquid level distribution information and the discrete liquid level distribution information at the previous moment, the current liquid level distribution status and the liquid level distribution status at the previous moment of each liquid level measurement point are obtained;
[0026] For each liquid level measurement point, if the current liquid level distribution state is true and the previous liquid level distribution state was false, then the current transfer vector is 1.
[0027] If the current liquid level distribution state is false and the previous liquid level distribution state was true, then the current transfer vector is -1.
[0028] If the current liquid level distribution state is false, and the previous liquid level distribution state was false, then the current transfer vector is 0;
[0029] If the current liquid level distribution state is true, and the previous liquid level distribution state was true, then the current transfer vector is 0;
[0030] Generate the current liquid level state transfer vector based on the current transfer vector of each liquid level measurement point.
[0031] Optionally, in a fourth implementation of the first aspect of the present invention, updating the preset dynamic model of the ink supply system based on the current ink cartridge measurement status, the current actual operating speed of the ink pump, the historical ink cartridge measurement status, and the historical actual operating speed of the ink pump includes:
[0032] The preset dynamic model of the ink supply system is the dynamic model of the ink supply system at the previous moment;
[0033] Input the current actual operating speed of the ink pump into the dynamic model of the ink supply system at the previous moment, and output the current predicted value;
[0034] Obtain historical forecast values;
[0035] The current loss function is calculated based on the preset delay order, the current ink cartridge measurement status, the current predicted value, the historical ink cartridge measurement status, and the historical predicted value.
[0036] Train the dynamic model of the ink supply system based on the current loss function, and calculate the gradient of the current loss function.
[0037] Obtain the neural network weight coefficients of the dynamic model of the ink supply system at the previous moment;
[0038] The weight coefficients of the current neural network are calculated based on the preset neural network learning rate, the neural network weight coefficients of the previous ink supply system dynamic model, and the current loss function gradient.
[0039] Based on the current weight coefficients of the neural network, the dynamic model of the ink supply system is updated to obtain the updated dynamic model of the ink supply system.
[0040] Optionally, in a fifth implementation of the first aspect of the present invention, the step of calculating the feedforward control quantity based on the updated dynamic model of the ink supply system, the current ink cartridge measurement status, the current actual operating speed of the ink pump, the historical ink cartridge measurement status, and the historical actual operating speed of the ink pump includes:
[0041] Based on the updated dynamic model of the ink supply system, the inverse solution function is obtained;
[0042] The preset target state of the ink supply cartridge, the current measurement state of the ink supply cartridge, the current actual operating speed of the ink pump, the historical measurement state of the ink supply cartridge, and the historical actual operating speed of the ink pump are input into the inverse solution function of the dynamic model of the ink supply system, and the feedforward control quantity is output.
[0043] A second aspect of the present invention provides an ink supply control system, comprising: a control device, an ink supply cartridge, an ink pump, a liquid level sensor, a pressure sensor, and a speed feedback sensor. The outlet of the ink pump is connected to the inlet of the ink supply cartridge. The liquid level sensor is used to acquire liquid level distribution information at at least two liquid level measurement points of the ink supply cartridge. The pressure sensor is used to acquire the liquid surface pressure of the ink supply cartridge. The speed feedback sensor is used to acquire the real-time speed of the ink pump. The control device is electrically connected to the ink pump, the liquid level sensor, the pressure sensor, and the speed feedback sensor, respectively. The control device is used to execute the ink supply control method described above.
[0044] Optionally, the liquid level sensor is mounted on the outside of the ink supply cartridge.
[0045] A third aspect of the present invention provides an ink supply control device, the ink supply control device comprising: a memory and at least one processor, wherein the memory stores instructions;
[0046] At least one of the processors invokes the instructions in the memory to cause the ink supply control device to perform the steps of the ink supply control method as described above.
[0047] A fourth aspect of the present invention provides a computer-readable storage medium storing instructions that, when executed by a processor, implement the steps of the ink supply control method described above.
[0048] In the technical solution of this invention, a dynamic model of the ink supply system is constructed. The dynamic model of the ink supply system is continuously updated by acquiring the measured status of the ink cartridge and the actual operating speed of the ink pump in real time. As online learning and training data increase, the feedforward control quantity is calculated through the dynamic model of the ink supply system, which has the ability to compensate for load changes, fluid characteristic fluctuations and pump wear. The ink pump control signal is calculated by the feedback control quantity and the feedforward control quantity together, which has strong control accuracy and anti-interference ability, and achieves high-precision and stable ink supply. Attached Figure Description
[0049] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which:
[0050] Figure 1 This is a first flowchart of the ink supply control method provided in an embodiment of the present invention;
[0051] Figure 2 This is a second flowchart of the ink supply control method provided in an embodiment of the present invention;
[0052] Figure 3 This is a third flowchart of the ink supply control method provided in the embodiments of the present invention;
[0053] Figure 4 This is a fourth flowchart of the ink supply control method provided in the embodiments of the present invention;
[0054] Figure 5 A fifth flowchart of the ink supply control method provided in an embodiment of the present invention;
[0055] Figure 6 A sixth flowchart of the ink supply control method provided in an embodiment of the present invention;
[0056] Figure 7 This is a flowchart illustrating the ink supply control method provided in an embodiment of the present invention.
[0057] Figure 8 This is a schematic diagram of the structure of the ink supply control device provided in an embodiment of the present invention. Detailed Implementation
[0058] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” or “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus 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 units not explicitly listed or inherent to such processes, methods, products, or apparatus.
[0059] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 and Figure 7 An embodiment of the ink supply control method in this invention is applied to an ink supply control system, which includes: a control device, an ink supply cartridge, and an ink pump. The outlet of the ink pump is connected to the inlet of the ink supply cartridge, and the control device is electrically connected to the ink pump.
[0060] The ink supply control method includes the following steps:
[0061] 101. Obtain the current ink cartridge measurement status and the current actual operating speed of the ink pump;
[0062] In this embodiment, the current... k At any given time, the ink cartridge is being measured. and the current actual operating speed of the ink pump .
[0063] In this embodiment, the real-time measurement status of the ink supply cartridge can include parameters such as air pressure, liquid level and temperature, and the actual operating speed of the ink pump at this moment can be collected. Through the above real-time data, the system can accurately reflect the dynamic changes in the current ink supply path, providing a basis for subsequent dynamic model updates of the ink supply system and ink pump control calculations.
[0064] 102. Obtain historical ink cartridge measurement status and historical ink pump actual operating speed;
[0065] The introduction of historical data is used to capture the hysteresis characteristics of the ink supply system, providing a data foundation for updating the dynamic model of the ink supply system.
[0066] 103. Update the preset dynamic model of the ink supply system based on the current ink cartridge measurement status, the current actual operating speed of the ink pump, the historical ink cartridge measurement status, and the historical actual operating speed of the ink pump.
[0067] In this embodiment, a preset dynamic model of the ink supply system is established. This dynamic model can employ a NARX model, first-order or multi-order difference equations, a linear state-space model, or other structures capable of describing the dynamic relationships of the ink supply system. The current ink cartridge measurement status, historical ink cartridge measurement status, current ink pump actual operating speed, and historical ink pump actual operating speed, obtained in steps 101 and 102, are input into the dynamic model. Figure 7 The middle part uses a neural network dynamic identifier to update the parameters of the ink supply system's dynamic model online, enabling the model to more accurately reflect the current actual equipment status. For example, the parameters of the ink supply system's dynamic model may change due to ink temperature, viscosity, ink supply tube stiffness, and pump aging; this step allows for dynamic tracking of these changes.
[0068] 104. Calculate the feedback control quantity based on the current ink supply cartridge measurement status, the preset ink supply cartridge target status, and the preset feedback coefficient;
[0069] In this embodiment, the feedback control quantity is calculated using the following formula:
[0070] Feedback control quantity =
[0071] in, The current measurement status of the ink cartridge. Preset target state for ink supply cartridge This is the current measurement status. With the target state Feedback coefficients for deviation design. Please refer to [link / reference]. Figure 7 The preset target state of the ink supply cartridge can be composed of a given constant air pressure and liquid level, which is a stable ink supply cartridge state.
[0072] The feedback control part is mainly used to quickly correct deviations and make the ink cartridge state converge towards the target value.
[0073] 105. Based on the updated dynamic model of the ink supply system, the current measurement status of the ink cartridge, the current actual operating speed of the ink pump, the historical measurement status of the ink cartridge, and the historical actual operating speed of the ink pump, the feedforward control quantity is calculated.
[0074] In this embodiment, after the dynamic model of the ink supply system is updated, the system predicts the trend of the ink cartridge state changing with the pump speed at the next moment. The compensation amount required to keep the ink cartridge state within the target range is inferred through the dynamic model of the ink supply system, and the feedforward control amount is obtained. The feedforward control amount is used to offset the pressure fluctuations that may occur in the system in advance.
[0075] 106. The ink pump control signal is calculated based on the feedback control quantity and the feedforward control quantity.
[0076] In this embodiment, the ink pump control signal is calculated as follows:
[0077]
[0078] in, This represents the ink pump control signal at time k+1. For feedback control, This is the feedforward control variable.
[0079] Please see Figure 7 , Figure 7 In this context, "controller" refers to the control device, and "motor" refers to the drive motor of the ink pump. The control device calculates the ink pump control signal and controls the drive motor of the ink pump to achieve stable ink supply.
[0080] In this embodiment, the final control signal of the ink pump is determined by both the feedback control quantity and the feedforward control quantity, which adjusts the ink pump speed.
[0081] In this embodiment of the invention, a dynamic model of the ink supply system is constructed. The dynamic model of the ink supply system is continuously updated by acquiring the measured status of the ink cartridge and the actual operating speed of the ink pump in real time. As online learning and training data increase, the feedforward control quantity is calculated through the dynamic model of the ink supply system, which has the ability to compensate for load changes, fluid characteristic fluctuations and pump wear. The ink pump control signal is calculated by the feedback control quantity and the feedforward control quantity together, which has strong control accuracy and anti-interference ability, and achieves high-precision and stable ink supply.
[0082] Please see Figure 2 One embodiment of the ink supply control method in this invention includes:
[0083] The ink supply control system further includes a liquid level sensor, a pressure sensor, and a speed feedback sensor. The liquid level sensor is used to acquire liquid level distribution information at at least two liquid level measurement points of the ink supply cartridge. The pressure sensor is used to acquire the liquid surface pressure of the ink supply cartridge. The speed feedback sensor is used to acquire the real-time speed of the ink pump. The control device is electrically connected to the liquid level sensor, the pressure sensor, and the speed feedback sensor.
[0084] 201. Control the liquid level sensor to acquire the liquid level information of each liquid level measurement point;
[0085] In this embodiment, there are at least two liquid level measurement points. The control device reads the liquid level information at each measurement point from the liquid level sensor to obtain the current liquid level at each measurement point. k Real-time liquid level information.
[0086] 202. Control the air pressure sensor to obtain the current liquid level air pressure;
[0087] In this embodiment, please refer to Figure 7 The control device obtains the first... k The air pressure at the ink cartridge level is constantly monitored and recorded as follows: ,Right now Figure 7 The ink is obtained through the ink supply system. The liquid level pressure is closely related to ink consumption, pump speed changes, and cartridge sealing, and is an important parameter characterizing the status of the ink supply cartridge.
[0088] 203. Control the speed feedback sensor to obtain the current actual operating speed of the ink pump;
[0089] In this embodiment, please refer to Figure 7 , Figure 7 The motor in the figure refers to the drive motor of the ink pump. The control device reads the current speed from the speed feedback sensor. k The actual operating speed of the ink pump at any given time is denoted as . By obtaining the actual operating speed of the ink pump, the dynamic model of the ink supply system can be trained, which helps to improve the accuracy of the model.
[0090] 204. Based on the current liquid level at each liquid level measurement point, obtain the current discrete liquid level distribution information;
[0091] Please see Figure 7 The control device constructs the first... based on the current liquid level information of each liquid level measuring point obtained in step 201. k The discrete distribution information of ink cartridge level at any given time is denoted as... ,Right now Figure 7 The ink is obtained from the ink supply system. This is used to characterize the actual shape of the liquid surface, such as: whether the ink is tilted forward or backward; whether the liquid level is in an abnormal drop or local collapse; and whether there are liquid surface fluctuations caused by high-speed nozzle spraying. It can convert the liquid level sequence of each measurement point into a function of the liquid surface state, and may include processing such as filtering, curve fitting, or difference analysis.
[0092] 204. Based on the current discrete liquid level distribution information and the discrete liquid level distribution information at the previous moment, obtain the current liquid level state transition vector;
[0093] In this embodiment, the discrete liquid level distribution information between two adjacent time points is further calculated to determine the current... k The liquid level state transition vector at time t is denoted as It is used to describe the dynamic characteristics of liquid level changes over time, such as: the rate of liquid level drop per unit time; the frequency and amplitude of liquid level fluctuations; and the trend of liquid level deformation caused by pump suction.
[0094] 205. Generate the current ink cartridge measurement status based on the current liquid level pressure and the current liquid level state transfer vector.
[0095] In this embodiment, by the firstk Liquid surface pressure at all times and current k Liquid level state transition vector at any moment , generate the first k Constantly measuring the status of the ink cartridge , .
[0096] In this embodiment of the invention, by setting up a pressure sensor, a speed feedback sensor, and multiple liquid level sensors, the current liquid level, liquid surface pressure, and actual operating speed of the ink pump at each liquid level measurement point of the ink cartridge are obtained. Based on the current liquid level at each measurement point, current discrete liquid level distribution information is generated. Then, using the discrete liquid level distribution information from the current and previous moments, a current liquid level state transition vector is generated. The current liquid level state transition vector and the current liquid surface pressure constitute the current ink cartridge measurement state, making the construction of the ink cartridge measurement state more complete and accurate. With the discrete liquid level distribution formed by multi-point liquid level measurements and the liquid level state transition vector, the system can not only accurately reflect the actual height and shape of the liquid surface but also capture the dynamic changes in the liquid surface over time, thereby identifying in advance liquid level fluctuations and pressure changes caused by printhead ejection, ink pump suction, or environmental disturbances. The fused state formed by combining liquid surface pressure data makes the update of the dynamic model closer to the real physical process, thereby improving the accuracy of feedforward prediction and feedback regulation. By relying on more accurate ink cartridge measurement status, the controller can adjust the ink pump operating speed more promptly, making the ink supply pressure and liquid level more stable, effectively reducing problems such as ink interruption, uneven spraying, and horizontal stripes, and significantly improving the steady-state performance, dynamic response capability and overall robustness of the ink supply system.
[0097] Please see Figure 3 One embodiment of the ink supply control method in this invention includes:
[0098] 301. For each liquid level measurement point, if the current liquid level is not lower than the measurement position, the current liquid level distribution status is marked as true;
[0099] 302. If the current liquid level is lower than the measurement position, the current liquid level distribution status is marked as false;
[0100] 303. Generate current discrete liquid level distribution information based on the current liquid level distribution status of each liquid level measurement point.
[0101] In this embodiment, discrete liquid level distribution information .
[0102] in, It is the first ink cartridge. k Time of the first i Liquid level distribution at each liquid level measurement point Indicates the ink supply cartridge number kTime of the first i The liquid level at each liquid level measuring point shall not be lower than the measuring position. Indicates the ink supply cartridge number k Time of the first i The liquid level at each liquid level measuring point is lower than the measuring position.
[0103] Please see Figure 7 Based on the status of the ink supply system and the current liquid level at each liquid level measurement point, the current discrete liquid level distribution information is obtained.
[0104] For each liquid level measurement point, if the sensor can acquire the liquid level height, the control device compares the current liquid level height fed back by the sensor with the height of the corresponding measurement position. If the sensor can only distinguish whether the liquid level is higher than the measurement position, the control device acquires the sensor signal and marks it as true or false. The measurement position can be adjusted by the installation position of the liquid level sensor. When the current liquid level at a measurement point is not lower than the measurement position, the system marks the liquid level distribution state of that measurement point as true, indicating that the measurement point is still covered by ink or its liquid level is within the normal range. If the liquid level height at a measurement point is lower than the measurement position, the liquid level distribution state of that measurement point is marked as false. This marking can be used to identify local drops, severe depressions, or low liquid level states. Discrete liquid level distribution information is used to represent the approximate shape of the ink cartridge liquid level, whether the liquid level distribution is uniform, and whether there are characteristics such as tilting, fluctuations, or local collapses of the liquid surface.
[0105] In this embodiment of the invention, through the aforementioned threshold-based liquid level status marking mechanism, this embodiment utilizes multiple liquid level measurement points to construct discrete liquid level distribution information composed of true or false values. This enables the ink supply system to obtain the distribution of liquid level in different spatial locations in real time with extremely low computational cost. By logically expressing whether each measurement point is covered by liquid, the system can quickly identify the overall height of the liquid level, local change trends, and potential anomalies, such as significant liquid level drops or deviations. This provides clear and stable basic data for subsequent liquid level dynamic analysis, liquid level state transition vector calculation, and ink cartridge measurement status construction. This discretization method not only improves the robustness and anti-interference capability of liquid level monitoring but also significantly enhances the response speed to liquid level changes. This allows the controller to more accurately assess the ink supply status and make timely control adjustments, thereby improving the stability and reliability of the entire ink supply system.
[0106] Please see Figure 4 One embodiment of the ink supply control method in this invention includes:
[0107] 401. Based on the current discrete liquid level distribution information and the discrete liquid level distribution information at the previous moment, obtain the current liquid level distribution status and the previous liquid level distribution status of each liquid level measurement point;
[0108] 402. For each liquid level measurement point, if the current liquid level distribution state is true and the previous liquid level distribution state was false, then the current transfer vector is 1.
[0109] 403. If the current liquid level distribution state is false and the previous liquid level distribution state was true, then the current transfer vector is -1.
[0110] 404. If the current liquid level distribution state is false, and the previous liquid level distribution state was false, then the current transfer vector is 0.
[0111] 405. If the current liquid level distribution state is true, and the previous liquid level distribution state was true, then the current transfer vector is 0;
[0112] 406. Generate the current liquid level state transfer vector based on the current transfer vector of each liquid level measurement point.
[0113] In this embodiment, the state transition vector is obtained according to the following table:
[0114] Table 1 State Transition Table
[0115]
[0116] In this embodiment of the invention, a state transition table is designed to perform state transitions on discrete liquid levels, transforming the continuous liquid level stability control problem into a distributed, state-invariant liquid level problem. For each liquid level measurement point, when the current liquid level distribution state is true and the previous state was false, it indicates that the liquid level at that location has experienced a rise, surface fluctuation, or replenishment within the current cycle, and the current transition vector of that measurement point is marked as 1. When the current liquid level distribution state is false and the previous state was true, it means that the liquid level at that measurement point has decreased, which may be a manifestation of a drop in liquid level, surface fluctuation, or excessively rapid ink pump suction, and the current transition vector of that measurement point is marked as −1. If a measurement point is false for two consecutive moments, it indicates that the location is continuously below the measurement position, and the liquid level has not changed significantly, and the transition vector is marked as 0. If a measurement point is true for two consecutive cycles, it indicates that the liquid level at that point remains stable and is continuously above the measurement position, and in this embodiment, the transition value of that measurement point is also marked as 0. Please refer to [link to relevant documentation]. Figure 7 Based on the discrete liquid level distribution information, the transfer vectors of all measurement points are obtained. Then, based on the state transition table, the first... k The liquid level state transition vector at time t is denoted as .
[0117] In this embodiment of the invention, the dynamic changes of the liquid level between two adjacent control cycles can be clearly and robustly characterized. It effectively distinguishes between three typical states: rising, falling, and no change in liquid level. Dynamic behaviors such as liquid level fluctuations, liquid level tilt compensation, and localized liquid level collapse caused by pump suction are quantified as discrete change signals of +1, 0, and -1. This makes the determination of liquid level change trends more stable, reduces computational load, and enhances noise immunity. With the help of this state transition vector, the control device can obtain more accurate information on liquid level change trends, thus providing a reliable basis for updating the dynamic model and feedforward prediction of the ink supply system, further improving the control accuracy and operational stability of the ink supply system.
[0118] Please see Figure 5 One embodiment of the ink supply control method in this invention includes:
[0119] 501. The preset dynamic model of the ink supply system is the dynamic model of the ink supply system at the previous moment;
[0120] 502. Input the current actual operating speed of the ink pump into the dynamic model of the ink supply system at the previous moment, and output the current predicted value;
[0121] 503. Obtain historical predicted values;
[0122] 504. Based on the preset delay order, the current ink cartridge measurement status, the current predicted value, the historical ink cartridge measurement status, and the historical predicted value, the current loss function is calculated.
[0123] 505. Train the dynamic model of the ink supply system based on the current loss function, and calculate the gradient of the current loss function;
[0124] 506. Obtain the neural network weight coefficients of the dynamic model of the ink supply system at the previous moment;
[0125] 507. Based on the preset neural network learning rate, the neural network weight coefficients of the previous ink supply system dynamic model, and the current loss function gradient, calculate the current neural network weight coefficients.
[0126] 508. Based on the current weight coefficients of the neural network, update the dynamic model of the ink supply system to obtain the dynamic update model of the ink supply system.
[0127] In this embodiment, the method for constructing the dynamic update model of the ink supply system is as follows:
[0128]
[0129] in, It is a dynamic update model for the ink supply system, and it is a nonlinear autoregressive neural network model. It is a neural network activation function Input, yes k The weight coefficients of the neural network at time step. The ink supply system dynamic update model is in k Always k The predicted output value of the ink cartridge measurement status at time +1.
[0130] exist middle, It is the output of the ink supply system's dynamic update model at a past moment. It is the delay order of the output state of the design. It is the actual input of the past in the dynamic update model of the ink supply system. It is the delay order of the control input in the design.
[0131] The neural network weight update rate is as follows:
[0132]
[0133] in, It is the learning rate of the designed neural network. It is the backpropagation loss function at time k. The gradient and loss function are designed as follows:
[0134]
[0135] The ink supply system dynamic model is trained via backpropagation based on the current loss function. The gradient of the loss function with respect to the weight coefficients of the neural network is calculated, and this gradient indicates the direction and magnitude of the adjustment needed for the current neural network weights. The weight coefficients are then written into the ink supply system dynamic model to complete the update of the ink supply system dynamic model for this cycle.
[0136] In this embodiment of the invention, through an online training and update mechanism for the dynamic model of the ink supply system, the dynamic model of the ink supply system can continuously adapt to changes in actual operating conditions during equipment operation, thereby significantly improving prediction accuracy and system stability. The model can capture the nonlinear characteristics, hysteresis characteristics, and dynamic response laws of the ink supply system in real time, and quickly identify the changing trends of ink cartridge level, air pressure, and flow rate over time. Through this online adaptive model update method, the ink supply system can obtain more accurate feedforward prediction capabilities, improve the calculation accuracy of ink pump control signals, reduce problems such as ink supply fluctuations, ink interruptions, and pressure oscillations, and improve the overall response speed, accuracy, and robustness of the ink supply system.
[0137] Please see Figure 6 One embodiment of the ink supply control method in this invention includes:
[0138] 601. Based on the updated dynamic model of the ink supply system, the inverse function is obtained;
[0139] 602. Input the preset target state of the ink supply cartridge, the current measurement state of the ink supply cartridge, the current actual operating speed of the ink pump, the historical measurement state of the ink supply cartridge, and the historical actual operating speed of the ink pump into the inverse solution function of the dynamic model of the ink supply system, and output the feedforward control quantity.
[0140] In this embodiment, It is based on the neural network dynamic identifier model from the target state The estimated feedforward control quantity, where, It is the inverse solution function for estimating the dynamic model, and its input is... for:
[0141]
[0142] in, This is the preset target state for the ink supply cartridge. , The preset target liquid level state transition vector. , For the preset first i The transfer vector of each liquid level measurement point The preset target liquid level pressure, It is the output of the ink supply system's dynamic update model at a past moment. It is the delay order of the output state of the design. It is the actual input of the past in the dynamic update model of the ink supply system. It is the delay order of the control input in the design.
[0143] Please refer to Figure 7 , Figure 7 The given pressure in the text represents the preset target liquid level pressure. Figure 7 The constant liquid level in the vector represents the preset target liquid level state transition vector. Figure 7 In express .
[0144] In this embodiment, based on the updated dynamic model of the ink supply system, its corresponding inverse function is constructed. This fully utilizes the dynamic characteristics of the ink supply system to achieve feedforward calculation of ink pump control commands. The preset target state of the ink cartridge, the current measured state of the ink cartridge, the current actual operating speed of the ink pump, and historical measured states and operating speeds of the ink cartridges are substituted into the inverse function to obtain the feedforward control quantity that meets the target state requirements. This quantity is then used to actively adjust the ink pump, achieving faster dynamic response and a more stable ink supply state.
[0145] The ink supply control method in the embodiments of the present invention has been described above. The ink supply control system in the embodiments of the present invention is described below. One embodiment of the ink supply control system in the embodiments of the present invention includes:
[0146] The system includes a control device, an ink supply cartridge, an ink pump, a liquid level sensor, a pressure sensor, and a speed feedback sensor. The outlet of the ink pump is connected to the inlet of the ink supply cartridge. The liquid level sensor is used to acquire liquid level distribution information at at least two liquid level measurement points of the ink supply cartridge. The pressure sensor is used to acquire the liquid surface pressure of the ink supply cartridge. The speed feedback sensor is used to acquire the real-time speed of the ink pump. The control device is electrically connected to the ink pump, the liquid level sensor, the pressure sensor, and the speed feedback sensor. The control device is used to execute the ink supply control method described above.
[0147] The ink supply control system in this embodiment is used to achieve real-time monitoring of changes in the liquid level inside the ink supply cartridge, online updating of the ink supply dynamic model, and precise control of the ink pump operation. Specifically, the ink supply cartridge outlet is connected to the printhead ink inlet, the ink pump outlet is connected to the ink supply cartridge inlet, and the ink pump supplies ink to the ink supply cartridge; a liquid level sensor is installed on the ink supply cartridge to collect liquid level height information at at least two liquid level measurement points in the ink supply cartridge; a pressure sensor is directly connected to the ink supply cartridge and directly installed on the ink supply cartridge to measure the liquid surface pressure of the ink supply cartridge; a speed feedback sensor is used to measure the real-time speed of the ink pump; the control device is electrically connected to the ink pump, liquid level sensor, pressure sensor, and speed feedback sensor respectively to acquire sensor data and output control commands to the ink pump. The ink supply control system may also include an air source and multiple shut-off valves. The air source is connected to the ink supply cartridge, the ink supply cartridge outlet is connected to the printhead ink inlet through a shut-off valve, the ink pump outlet is connected to the ink supply cartridge inlet through a shut-off valve, and the air pressure sensor is directly connected to the air source and the ink supply cartridge and is directly installed on the ink supply cartridge to measure the liquid level air pressure of the ink supply cartridge.
[0148] In one embodiment of the ink supply control system of the present invention, the liquid level sensor is installed on the outside of the ink supply cartridge.
[0149] In this embodiment, non-invasive detection of the liquid level inside the ink cartridge is achieved, effectively avoiding the leakage risks, limited installation space, and impact on the liquid that may be caused by traditional internal liquid level sensors. Simultaneously, the external installation method improves the convenience of later maintenance and replacement. The non-contact liquid level sensor is installed on the outside of the ink cartridge to obtain the current liquid level distribution information and includes at least two measurement points. The liquid level sensor can be an ultrasonic liquid level sensor, an optical liquid level sensor, a capacitive liquid level sensor, or other non-contact liquid level detection device for external measurement of the liquid level inside the ink cartridge. By installing the liquid level sensor on the outside of the ink cartridge, additional space is avoided inside the ink cartridge, eliminating the need for structural openings in the ink cartridge body, improving the sealing reliability of the ink cartridge, and solving the problem in existing technologies where sensors are easily damaged or contaminated when in contact with corrosive or high-viscosity fluids.
[0150] In this embodiment, traditional flow meters and contact pressure sensors are abandoned, and the dynamics of the system can be accurately characterized using only non-contact liquid level information, air pressure data, and pump speed commands.
[0151] Figure 8 This is a schematic diagram of the structure of an ink supply control device 700 provided in an embodiment of the present invention. The ink supply control device 700 can vary significantly due to different configurations or performance characteristics. It may include one or more central processing units (CPUs) 710 (e.g., one or more processors) and a memory 720, and one or more storage media 730 (e.g., one or more mass storage devices) storing application programs 733 or data 732. The memory 720 and storage media 730 can be temporary or persistent storage. The program stored in the storage media 730 may include one or more modules (not shown in the diagram), each module including a series of instruction operations on the ink supply control device 700. Furthermore, the processor 710 may be configured to communicate with the storage media 730 and execute the series of instruction operations in the storage media 730 on the ink supply control device 700 to implement the steps of the ink supply control methods provided in the above-described method embodiments.
[0152] The ink supply control device 700 may also include one or more power supplies 740, one or more wired or wireless network interfaces 750, one or more input / output interfaces 760, and / or one or more operating systems 731, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will understand that... Figure 8The illustrated ink supply control device structure does not constitute a limitation on the ink supply control device, and may include more or fewer components than illustrated, or combine certain components, or have different component arrangements.
[0153] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to perform the steps of the ink supply control method.
[0154] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the system, device, or unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.
[0155] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0156] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
[0157] The preferred embodiments of the present invention have been described in detail above, but the present disclosure is not limited to the embodiments described. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention, and these equivalent modifications or substitutions are all included within the scope defined by the claims of the present disclosure.
Claims
1. An ink supply control method characterized by, The application is applied to an ink supply control system, which comprises a control device, an ink supply box and an ink pump, the outlet of the ink pump is connected with the inlet of the ink supply box, and the control device is electrically connected with the ink pump. The ink supply control method comprises the following steps: acquiring a current ink supply box measurement state and a current ink pump actual running speed; acquiring a historical ink supply box measurement state and a historical ink pump actual running speed; updating a preset ink supply system dynamic model according to the current ink supply box measurement state, the current ink pump actual running speed, the historical ink supply box measurement state and the historical ink pump actual running speed; calculating a feedback control amount according to the current ink supply box measurement state, a preset ink supply box target state and a preset feedback coefficient; calculating a feedforward control amount according to the updated ink supply system dynamic model, the current ink supply box measurement state, the current ink pump actual running speed, the historical ink supply box measurement state and the historical ink pump actual running speed; calculating an ink pump control signal according to the feedback control amount and the feedforward control amount; the updating of the preset ink supply system dynamic model according to the current ink supply box measurement state, the current ink pump actual running speed, the historical ink supply box measurement state and the historical ink pump actual running speed comprises: the preset ink supply system dynamic model is an ink supply system dynamic model at a previous moment; inputting the current ink pump actual running speed into the ink supply system dynamic model at the previous moment to output a current prediction value; acquiring a historical prediction value; calculating a current loss function according to a preset delay order, the current ink supply box measurement state, the current prediction value, the historical ink supply box measurement state and the historical prediction value; training the ink supply system dynamic model according to the current loss function to calculate a current loss function gradient; acquiring a neural network weight coefficient of the ink supply system dynamic model at the previous moment; calculating a current neural network weight coefficient according to a preset neural network learning rate, the neural network weight coefficient of the ink supply system dynamic model at the previous moment and the current loss function gradient; updating the ink supply system dynamic model according to the current neural network weight coefficient to obtain an updated ink supply system dynamic model; the calculating of the feedforward control amount according to the updated ink supply system dynamic model, the current ink supply box measurement state, the current ink pump actual running speed, the historical ink supply box measurement state and the historical ink pump actual running speed comprises: obtaining an inverse solution function according to the updated ink supply system dynamic model; inputting a preset ink supply box target state, the current ink supply box measurement state, the current ink pump actual running speed, the historical ink supply box measurement state and the historical ink pump actual running speed into the inverse solution function of the ink supply system dynamic model to output the feedforward control amount.
2. The ink supply control method according to claim 1, characterized by: The ink supply control system further comprises a liquid level sensor, an air pressure sensor and a rotation speed feedback sensor, the liquid level sensor is used to obtain liquid level distribution information of at least two liquid level measuring points of the ink supply box, the air pressure sensor is used to obtain liquid surface air pressure of the ink supply box, and the rotation speed feedback sensor is used to obtain real-time rotation speed of the ink pump, and the control device is electrically connected with the liquid level sensor, the air pressure sensor and the rotation speed feedback sensor; the current ink supply box measuring state and the current ink pump actual running speed are obtained, including: controlling the liquid level sensor to obtain the current liquid level of each liquid level measuring point; controlling the air pressure sensor to obtain the current liquid surface air pressure; controlling the rotation speed feedback sensor to obtain the current ink pump actual running speed; obtaining current discrete liquid level distribution information according to the current liquid level of each liquid level measuring point; obtaining a current liquid level state transition vector according to the current discrete liquid level distribution information and the last time discrete liquid level distribution information; generating a current ink supply box measuring state according to the current liquid surface air pressure and the current liquid level state transition vector.
3. The ink supply control method according to claim 2, characterized by: The current discrete liquid level distribution information is obtained according to the current liquid level of each liquid level measuring point, including: for each liquid level measuring point, if the current liquid level is not lower than the measuring position, the current liquid level distribution state is marked as true; if the current liquid level is lower than the measuring position, the current liquid level distribution state is marked as false; generating the current discrete liquid level distribution information according to the current liquid level distribution state of each liquid level measuring point.
4. The ink supply control method according to claim 3, characterized by: The current liquid level state transition vector is obtained according to the current discrete liquid level distribution information and the last time discrete liquid level distribution information, including: obtaining the current liquid level distribution state and the last time liquid level distribution state of each liquid level measuring point according to the current discrete liquid level distribution information and the last time discrete liquid level distribution information; for each liquid level measuring point, if the current liquid level distribution state is true and the last time liquid level distribution state is false, the current transition vector is 1; if the current liquid level distribution state is false and the last time liquid level distribution state is true, the current transition vector is -1; if the current liquid level distribution state is false and the last time liquid level distribution state is false, the current transition vector is 0; if the current liquid level distribution state is true and the last time liquid level distribution state is true, the current transition vector is 0; generating the current liquid level state transition vector according to the current transition vector of each liquid level measuring point.
5. An ink supply control system characterized by comprising: The ink supply control system comprises a control device, an ink supply box, an ink pump, a liquid level sensor, an air pressure sensor and a rotation speed feedback sensor, the outlet of the ink pump is connected with the inlet of the ink supply box, the liquid level sensor is used to obtain liquid level distribution information of at least two liquid level measuring points of the ink supply box, the air pressure sensor is used to obtain liquid surface air pressure of the ink supply box, and the rotation speed feedback sensor is used to obtain real-time rotation speed of the ink pump, and the control device is electrically connected with the ink pump, the liquid level sensor, the air pressure sensor and the rotation speed feedback sensor respectively; the control device is used to execute the ink supply control method in any one of claims 1-4.
6. The ink supply control system of claim 5, wherein: The liquid level sensor is installed outside the ink supply cartridge.
7. An ink supply control device characterized by comprising: The ink supply control device comprises a memory and at least one processor, and the memory stores instructions; The at least one processor invokes the instructions in the memory to enable the ink supply control device to perform the steps of the ink supply control method according to any one of claims 1-4.
8. A computer-readable storage medium having stored thereon instructions, the computer-readable storage medium comprising: The instructions, when executed by a processor, implement the steps of the ink supply control method according to any one of claims 1-4.
Citation Information
Patent Citations
Nonlinear dynamic compensation adjustment method for piezoelectric ceramic actuator
CN119882405A
Ink Flow Regulation Mechanism
US20120206515A1