Inkjet waveform optimization method, apparatus, device, and storage medium

Abstract

The present application relates to the field of inkjet printing technology, and more particularly to an inkjet waveform optimization method, device, equipment and storage medium, the method is through the deep integration of dynamic closed loop feedback mechanism and adaptive multi-pulse driving strategy by using the tentacle algorithm, which significantly improves the precision, stability and adaptability of inkjet printing; in terms of printing accuracy, the volume and speed deviation of ink droplets are controlled in a very small range, effectively inhibiting satellite droplet and tailing phenomenon, which can meet the demand of micron-level high-precision scene; in terms of optimization efficiency, compared with the traditional artificial trial and error method, the debugging cycle is significantly reduced, and the waste of ink and substrate is reduced; in terms of dynamic adaptability, it can respond to dynamic interference such as ink viscosity change, nozzle aging, environmental temperature and humidity fluctuation in real time, and automatically adjust the waveform parameters to ensure the stability of continuous printing process; in terms of energy consumption and hardware life, the invalid energy consumption is reduced through fine waveform control, the aging speed of piezoelectric element is slowed down, and the service life of hardware is significantly prolonged.

Description

Technical Field

[0001] This invention relates to the field of inkjet printing technology, and in particular to an inkjet waveform optimization method, apparatus, device, and storage medium. Background Technology

[0002] Piezoelectric ceramic inkjet printing technology utilizes the inverse piezoelectric effect of piezoelectric ceramic materials to achieve high-precision ink droplet control. Its core principle is: applying a specific voltage waveform to the piezoelectric ceramic element causes the piezoelectric material to deform and squeeze the ink cavity, generating instantaneous pressure to drive the ink droplet ejection.

[0003] Piezoelectric ceramic inkjet printing technology is widely used in the electronics industry (such as OLED display manufacturing), ceramic decorative printing, and biomedicine (such as cell printing) due to its advantages such as no need to heat the ink, long printhead life, and support for high-resolution printing. Compared with thermal inkjet technology, piezoelectric ceramic inkjet technology effectively avoids the problem of ink thermal degradation and is compatible with high-viscosity and high-solids-content functional inks. Moreover, through fine adjustment of waveform parameters, it can precisely control the volume, shape, and trajectory of ink droplets to meet the printing needs of high-precision patterns such as micron-level circuits.

[0004] However, the parameter design of the inkjet drive waveform is the core technical bottleneck restricting the improvement of print quality. The drive waveform is composed of parameters such as voltage amplitude, pulse width, rise / fall slope, and multi-pulse timing. The combination of these parameters directly determines the pressure wave propagation characteristics in the ink cavity, the ink droplet formation process, and the jetting stability. Existing waveform optimization methods have many obvious defects.

[0005] Traditional methods rely on manual experience or fixed parameter templates for trial-and-error adjustments, which involves repeatedly modifying parameters such as the rise time and peak voltage of trapezoidal or square waves and verifying the effect through printing tests. This approach is inefficient and struggles to cope with dynamic conditions such as changes in ink viscosity and printhead temperature fluctuations. Especially in scenarios where multiple printheads work together, the complexity of parameter adjustments increases exponentially, leading to extended debugging cycles and significant material waste.

[0006] Some improved technologies attempt to introduce static feedback mechanisms, such as capturing the ink droplet ejection process with a high-speed camera, extracting feature parameters such as droplet size, trail length, or number of satellite dots using image processing algorithms, and then adjusting the drive waveform based on a preset threshold. However, such methods suffer from feedback delays and algorithmic limitations. Static feedback only optimizes the results of a single ejection and cannot respond in real time to dynamic interferences such as ink pressure fluctuations and printhead mechanical fatigue during continuous printing. At the same time, traditional image processing algorithms are susceptible to motion blur or noise interference under high-speed shooting conditions, and the accuracy of droplet edge recognition is insufficient, making it difficult to support high-resolution printing requirements.

[0007] At the level of drive waveform design, existing technologies mostly adopt single-pulse or fixed-timing multi-pulse drive modes. Single-pulse drive (such as trapezoidal wave) drives the deformation of piezoelectric ceramic through a single voltage pulse. The structure is simple, but it is difficult to suppress the reflection of pressure waves in the ink cavity, which can easily cause ink droplet splitting or satellite point phenomenon, resulting in blurred edges of printed patterns. Although fixed-timing multi-pulse drive technology can improve the jetting stability of low viscosity ink by applying pressure multiple times, its pulse interval and amplitude are usually fixed parameters, which cannot adapt to different ink characteristics. Moreover, multi-pulse drive requires higher voltage support, which will increase system energy consumption and accelerate the aging of piezoelectric elements.

[0008] Existing optimization methods generally lack intelligent adaptability to complex nonlinear systems: traditional parameter search methods are prone to getting trapped in local optima and are difficult to efficiently cover multidimensional parameter spaces; at the same time, existing methods do not fully consider the coupling effect of printhead state drift (such as the attenuation of piezoelectric ceramic response due to long-term use) and ink characteristics, thus limiting the universality of optimization results.

[0009] The aforementioned problems are particularly prominent in high-precision, high-speed printing scenarios. For example, in the metal mesh printing process of OLED displays, deviations in the volume of satellite dots or ink droplets may cause short circuits or open circuits, directly reducing product yield.

[0010] Compared with existing related patents: CN118061671A monitors the nozzle throat pressure in real time using a SAW pressure sensor and dynamically adjusts the input voltage using model predictive control (MPC). While this solves the problem of traditional methods failing to track printhead state drift in real time, it relies on the installation accuracy of the SAW sensor and does not consider the impact of dynamic changes in ink viscosity on the mapping relationship between pressure fluctuations and droplet volume. The MPC parameters are fixed and lack an adaptive mechanism, potentially leading to delayed closed-loop control response under complex operating conditions. CN118818989A integrates multi-dimensional parameters through a dynamic reward function and constructs an inkjet dynamics model based on droplet motion equations, overcoming the limitations of fixed rules in traditional PID control. However, the reinforcement learning model relies on high-precision sensor data, making it susceptible to noise interference in real-world industrial scenarios. Furthermore, the high computational complexity of the droplet trajectory equations makes it difficult to meet the real-time requirements of high-speed coding. The 118567590A integrates the Bellman-Ford algorithm to optimize the printing path and dynamically adjusts the ink volume based on color saturation. However, the congestion coefficient in the path optimization algorithm relies on manual experience and presets, and the drying time assessment does not consider dynamic changes in ambient temperature and humidity, which may lead to the failure of paper feed speed optimization. The CN118544710A quantifies the entropy value of heat energy distribution through an improved Shannon entropy algorithm and dynamically adjusts the power of the water chiller by combining Kalman filtering, avoiding the energy waste of traditional temperature control methods. However, the weighting coefficients and smoothing parameters in the entropy calculation need to be calibrated offline, and the water valve adjustment is prone to lag under high-frequency control. The CN118003770A uses LVDS differential signal transmission control commands to solve the signal distortion problem under long-distance transmission. However, the hardware modification cost of the differential drive circuit is high, the inkjet intensity control relies on a fixed voltage pulse template, and no adaptive drive strategy is designed for different ink viscosities.

[0011] In summary, the main shortcomings of existing piezoelectric inkjet waveform optimization technologies can be summarized as follows: insufficient dynamic adaptability, low optimization efficiency, limited droplet control accuracy, and high system energy consumption and hardware costs; therefore, there is still room for improvement in existing technologies. Summary of the Invention

[0012] To overcome the shortcomings of existing technologies, the present invention aims to provide an inkjet waveform optimization method that significantly improves the accuracy, stability and adaptability of inkjet printing through the deep integration of the beetle whisker algorithm, dynamic closed-loop feedback mechanism and adaptive multi-pulse drive strategy.

[0013] The first aspect of this invention provides an inkjet waveform optimization method. The inkjet testing system includes an industrial control computer and a piezoelectric inkjet nozzle, a waveform driver, and a droplet observer, all electrically connected to the industrial control computer. The industrial control computer outputs an initial waveform or an optimized waveform to the piezoelectric inkjet nozzle via the waveform driver to trigger droplet ejection. The droplet observer is used to acquire real-time droplet information. The inkjet waveform optimization method includes: acquiring target parameters, iteration parameters, and ink characteristics; generating an initial waveform based on the ink characteristics and initializing the iteration parameters; outputting the initial waveform to trigger droplet ejection and acquiring real-time... Ink droplet information; based on the target parameters, the real-time ink droplet information, and the initialized iteration parameters, an optimized waveform is generated using the beetle whisker algorithm; the optimized waveform is output to trigger ink droplet ejection, and real-time optimized ink droplet information is obtained; a preset deviation threshold is obtained, and a state deviation problem is determined based on the real-time optimized ink droplet information and the preset deviation threshold; if a state deviation problem exists, the real-time ink droplet information is replaced with the real-time optimized ink droplet information, and the process returns to execute the optimization waveform generation using the beetle whisker algorithm based on the target parameters, the real-time ink droplet information, and the initialized iteration parameters.

[0014] Optionally, in a first implementation of the first aspect of the present invention, the step of obtaining target parameters, iteration parameters, and ink characteristics, generating an initial waveform based on the ink characteristics, and initializing the iteration parameters includes: obtaining target parameters and ink characteristics, wherein the target parameters include target droplet velocity and target droplet volume, and the ink characteristics include current ink viscosity and current ink surface tension; obtaining iteration parameters, wherein the iteration parameters include interparticle spacing, interparticle attenuation factor, search step size, step size attenuation factor, random direction vector, and minimum value to avoid divergence; initializing the interparticle spacing and the search step size to obtain initial interparticle spacing and initial search step size; and randomly generating 256 voltage points based on the current ink viscosity to form the initial waveform.

[0015] Optionally, in a second implementation of the first aspect of the present invention, the step of outputting the initial waveform to trigger ink droplet ejection and obtaining real-time ink droplet information includes: converting the initial waveform into initial analog driving information and outputting it to the piezoelectric inkjet nozzle through a waveform driver to trigger ink droplet ejection; obtaining a real-time continuous image of ink droplets through an ink droplet observer; and sequentially performing preprocessing, segmentation processing, and feature extraction processing on the real-time continuous image of ink droplets to obtain real-time ink droplet information.

[0016] Optionally, in a third implementation of the first aspect of the present invention, the step of sequentially performing preprocessing, segmentation, and feature extraction processing on the real-time continuous ink droplet image to obtain real-time ink droplet information includes: sequentially performing motion blur restoration processing, noise suppression processing, and contrast enhancement processing on the real-time continuous ink droplet image to complete preprocessing and obtain a preprocessed ink droplet image; sequentially performing adaptive threshold segmentation processing and edge correction and optimization processing on the preprocessed ink droplet image to complete segmentation and obtain an ink droplet contour positioning image; obtaining the real-time ink droplet volume from the ink droplet contour positioning image using contour integration and equivalent sphere method, obtaining the real-time ink droplet velocity using inter-frame template matching and sub-pixel positioning algorithm, and extracting ink droplet morphology auxiliary features; determining the ink droplet category based on the ink droplet morphology auxiliary features using a regularized threshold judgment algorithm; and integrating the ink droplet category, the real-time ink droplet velocity, and the real-time ink droplet volume to obtain real-time ink droplet information.

[0017] Optionally, in the fourth implementation of the first aspect of the present invention, the step of generating an optimized waveform using the beetle whisker algorithm based on the target parameters, the real-time ink droplet information, and the initialized iteration parameters includes: substituting the target parameters and the real-time ink droplet information into a pre-constructed mapping quality index formula to obtain the current ejection quality index; determining whether a preset termination search condition is met based on the current ejection quality index; if not met, generating an optimized waveform using the beetle whisker algorithm based on the initialized iteration parameters; if met, stopping the generation of the optimized waveform.

[0018] Optionally, in the fifth implementation of the first aspect of the present invention, the step of generating an optimized waveform using the beetle whisker algorithm based on the initialized iteration parameters includes: calculating the left and right antenna waveforms based on the initial waveform and the random direction vector; calculating the left and right fitness values ​​based on the left and right antenna waveforms; updating the initial waveform according to the left and right fitness values ​​to obtain an initial optimized waveform; updating the initial antenna spacing and the initial search step size to obtain an updated antenna spacing and an updated search step size; determining whether a preset iteration search stopping condition is met based on the updated antenna spacing and the updated search step size, and determining whether a preset termination search condition is met based on the left and right fitness values; if either condition is not met, the initial waveform is replaced with the initial optimized waveform, and the step of calculating the left and right antenna waveforms based on the initial waveform and the random direction vector is returned to execution; if both conditions are met, the optimized waveform is output.

[0019] Optionally, in a sixth implementation of the first aspect of the present invention, obtaining a preset deviation threshold and determining whether a state deviation problem exists based on the real-time optimized ink droplet information and the preset deviation threshold includes: obtaining the preset deviation threshold; inputting the real-time optimized ink droplet information and the target parameter into a pre-constructed mapping quality index formula to obtain an optimized jetting quality index; and comparing the optimized jetting quality index with the preset deviation threshold to determine whether a state deviation problem exists.

[0020] A second aspect of the present invention provides an inkjet waveform optimization device, comprising: an acquisition module for acquiring target parameters, iteration parameters, and ink characteristics, generating an initial waveform based on the ink characteristics, and initializing the iteration parameters; a first output module for outputting the initial waveform to trigger ink droplet ejection and acquiring real-time ink droplet information; a first iteration module for generating an optimized waveform using a beetle whisker algorithm based on the target parameters, the real-time ink droplet information, and the initialized iteration parameters; a second output module for outputting the optimized waveform to trigger ink droplet ejection and acquiring real-time optimized ink droplet information; a judgment module for acquiring a preset deviation threshold and judging whether a state deviation problem exists based on the real-time optimized ink droplet information and the preset deviation threshold; and a second iteration module for replacing the real-time ink droplet information with the real-time optimized ink droplet information if a state deviation problem exists, and returning to execute the generation of the optimized waveform using the beetle whisker algorithm based on the target parameters, the real-time ink droplet information, and the initialized iteration parameters.

[0021] A third aspect of the present invention provides an inkjet waveform optimization device, the inkjet waveform optimization device comprising: a memory and at least one processor, the memory storing instructions; the at least one processor calling the instructions in the memory to cause the inkjet waveform optimization device to perform the various steps of the inkjet waveform optimization method described in any of the preceding claims.

[0022] 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 inkjet waveform optimization method described in any of the preceding claims.

[0023] The technical solution of this invention significantly improves the accuracy, stability, and adaptability of inkjet printing through the deep integration of the beetle whisker algorithm, dynamic closed-loop feedback mechanism, and adaptive multi-pulse drive strategy. Regarding printing accuracy, droplet volume and velocity deviations are controlled within an extremely small range, effectively suppressing satellite droplets and trailing phenomena, resulting in clear printed pattern edges. This meets the micron-level high-precision requirements of scenarios such as metal mesh printing for OLED displays, significantly improving product yield. In terms of optimization efficiency, the iteration optimization cycle is greatly shortened, significantly reducing the debugging cycle compared to traditional manual trial-and-error methods, thus reducing ink and substrate waste. Regarding dynamic adaptability, it can respond in real time to dynamic interferences such as changes in ink viscosity, printhead aging, and fluctuations in ambient temperature and humidity, automatically adjusting waveform parameters without manual intervention to ensure stable continuous printing. Regarding energy consumption and hardware lifespan, refined waveform control reduces ineffective energy consumption, slows down the aging rate of piezoelectric components, significantly extends hardware lifespan, and reduces system operating costs. In terms of compatibility, it can adapt to various functional inks with high viscosity and high solids content, is compatible with existing piezoelectric inkjet printing system hardware architectures, requires no large-scale hardware modifications, and has broad application scenarios and promotional value. Attached Figure Description

[0024] Figure 1 A logic flowchart of the inkjet waveform optimization method provided in an embodiment of the present invention;

[0025] Figure 2 This is a schematic diagram of the structure of the inkjet waveform optimization device provided in an embodiment of the present invention;

[0026] Figure 3 This is a schematic diagram of the structure of the inkjet waveform optimization device provided in an embodiment of the present invention. Detailed Implementation

[0027] This invention provides an inkjet waveform optimization method, apparatus, device, and storage medium. In this invention, the terms "first," "second," "third," "fourth," etc. (if present)," in the specification, claims, and accompanying drawings 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 a sequence other than that 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.

[0028] The inkjet testing system of this invention provides a physical platform for inkjet waveform optimization methods. The system includes an industrial control computer and piezoelectric inkjet nozzles, a waveform driver, and a droplet observer, all electrically connected to the computer. The droplet observer consists of a nano-pulse stroboscopic light source and a CCD camera. Through the coordinated operation of these components, the system achieves a basic closed loop encompassing waveform output, droplet ejection, status monitoring, and data feedback. The piezoelectric inkjet nozzle employs a squeeze-mode design, with a piezoelectric sensor attached to the outside of a glass capillary. After receiving a voltage waveform signal, the nozzle squeezes the ink chamber through the inverse piezoelectric effect to eject droplets. The waveform driver executes the optimized waveform parameters, outputs precise voltage pulses, supports multi-pulse timing adjustment, and matches waveform frequencies from 0 to 30 kHz. The nano-pulse stroboscopic light source provides high-frequency, low-interference illumination for droplet observation; its pulse width is adapted to high-speed imaging to avoid motion blur, and it is triggered synchronously with the printhead ejection action. The ink droplet observation instrument integrates a CCD camera and an image processing module, capable of capturing ink droplet ejection images and extracting key feature parameters, including droplet type, velocity, and volume. The industrial control computer, as the core control unit, runs control programs and optimization algorithms, processes observation data, and outputs control commands, supporting real-time data transmission and parallel computing. Specifically, the industrial control computer outputs an initial or optimized waveform to the piezoelectric inkjet nozzle via a waveform driver to trigger ink droplet ejection. The nano-pulse stroboscopic light source works synchronously with the ejection action, providing clear illumination for ink droplet observation. The ink droplet observation instrument captures the ejection process in real time, extracts parameters such as droplet type, velocity, and volume, and transmits them back to the industrial control computer. The industrial control computer compares the measured parameters with the target parameters, adjusts the waveform parameters through optimization algorithms, and then outputs the waveform to the piezoelectric inkjet nozzle via the waveform driver, forming a basic data link at the hardware level, thus providing physical support for subsequent closed-loop optimization.

[0029] This application discloses an inkjet waveform optimization method. For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 One embodiment of the inkjet waveform optimization method in this invention includes:

[0030] 101. Obtain the target parameters, iteration parameters, and ink characteristics; generate an initial waveform based on the ink characteristics; and initialize the iteration parameters.

[0031] In this embodiment, the industrial control computer first obtains the target parameters corresponding to the printing task, the iteration parameters required by the beetle whisker algorithm, and the core characteristic parameters of the ink currently being printed. The target parameters are precisely defined according to the printing scenario, for example, the target ink droplet volume for printing a metal mesh on an OLED display. Target droplet velocity ;

[0032] The initialized iteration parameters include the initial antennal spacing, antennal attenuation factor, initial search step size, step size attenuation factor, random direction vector, and minimum value to avoid divergence; the initial antennal spacing Interantennae attenuation factor That is, after each iteration, the antennal spacing is updated to... The initial search step size The step size attenuation factor is set to 8% of the voltage parameter range (0~130V), i.e., 10.4V. After each iteration, update the search step size to... The minimum value to avoid divergence This is used to prevent numerical divergence during algorithm calculation; the random direction vector It is a 256-dimensional vector.

[0033] 102. Output the initial waveform to trigger ink droplet ejection and acquire real-time ink droplet information;

[0034] In this embodiment, the industrial control computer uses a waveform drive control program to generate the initial waveform. The signal is converted into an analog drive signal, amplified by a waveform driver, and output to the piezoelectric inkjet nozzle. This drives the piezoelectric ceramic element to deform and squeeze the ink chamber, triggering the ejection of ink droplets. Simultaneously, a nano-pulse stroboscopic light source and a CCD camera are activated. Based on the stroboscopic principle, single-flash high-speed imaging is achieved, capturing continuous dynamic images of the ink droplets as they leave the nozzle and fly through the air. After processing the continuous dynamic images, real-time ink droplet information is obtained.

[0035] 103. Based on the target parameters, the real-time ink droplet information, and the initialized iteration parameters, the longhorn beetle whisker algorithm is used to generate an optimized waveform;

[0036] In this embodiment, image processing and feature extraction are performed on the real-time raw ink droplet data (continuous dynamic image) to obtain data including ink droplet categories. Real-time droplet speed Real-time droplet volume The real-time ink droplet information is obtained; combined with the initialized iterative parameters, the beetle whisker algorithm is called to perform a multi-dimensional parameter space search on the initial waveform to generate an optimized waveform.

[0037] 104. Output the optimized waveform to trigger ink droplet ejection and obtain real-time optimized ink droplet information;

[0038] In this embodiment, the optimized waveform is output to the piezoelectric inkjet nozzle through a waveform driver to trigger ink droplet ejection; the ink droplet observation and data processing process is repeated to collect the optimized real-time ink droplet information for quantitative verification of the optimization effect.

[0039] 105. Obtain a preset deviation threshold, and determine whether there is a state deviation problem based on the real-time optimized ink droplet information and the preset deviation threshold;

[0040] In this embodiment, the preset is based on the injection index. Deviation threshold, such as in high-precision scenarios Medium precision scenarios Low-precision scenes The real-time optimized droplet information is substituted into the pre-constructed jetting index formula to calculate the optimized jetting quality index. The index is then compared with the preset deviation threshold to determine whether the current optimized waveform has deviated from its state due to dynamic operating conditions, such as ink viscosity drift, printhead temperature fluctuation, piezoelectric element response decay, etc.

[0041] 106. If there is a state deviation problem, the real-time ink drop information is replaced with the real-time optimized ink drop information, and the process is returned to execute the optimization waveform generated by the beetle whisker algorithm based on the target parameters, the real-time ink drop information and the initialized iteration parameters;

[0042] In this embodiment, if there is a state deviation, i.e. the optimized jetting index does not reach the deviation threshold, the real-time optimized ink droplet information is replaced with the original real-time ink droplet information, and the intelligent optimization stage is returned to re-execute the beetle whisker algorithm optimization; if there is no state deviation, the optimization process is terminated and the final optimized waveform is output.

[0043] The inkjet waveform optimization method disclosed in this application significantly improves the accuracy, stability, and adaptability of inkjet printing through the deep integration of the beetle whisker algorithm, dynamic closed-loop feedback mechanism, and adaptive multi-pulse driving strategy. Regarding printing accuracy, droplet volume and velocity deviations are controlled within an extremely small range, effectively suppressing satellite droplets and trailing phenomena, resulting in clear printed pattern edges. This meets the micron-level high-precision requirements of scenarios such as metal mesh printing for OLED displays, significantly improving product yield. In terms of optimization efficiency, the iterative optimization cycle is greatly shortened. Compared to traditional manual trial-and-error methods, the debugging cycle is significantly reduced, lowering ink and... It reduces substrate waste; in terms of dynamic adaptability, it can respond in real time to dynamic disturbances such as changes in ink viscosity, printhead aging, and fluctuations in ambient temperature and humidity, automatically adjusting waveform parameters without manual intervention to ensure stable continuous printing; in terms of energy consumption and hardware lifespan, it reduces ineffective energy consumption through refined waveform control, slows down the aging rate of piezoelectric components, significantly extends hardware lifespan, and reduces system operating costs; in terms of compatibility, it can adapt to various functional inks such as high viscosity and high solids content, and is compatible with existing piezoelectric inkjet printing system hardware architectures without large-scale hardware modifications, possessing broad application scenarios and promotional value.

[0044] Furthermore, in this embodiment of the invention, the steps of obtaining the target parameters, iteration parameters, and ink characteristics, generating an initial waveform based on the ink characteristics, and initializing the iteration parameters include:

[0045] 201. Obtain target parameters and ink characteristics, wherein the target parameters include target droplet velocity and target droplet volume, and the ink characteristics include current ink viscosity and current ink surface tension;

[0046] In this embodiment, the target parameter focuses on key indicators of printing accuracy and is specifically defined as the target ink droplet velocity. and target ink droplet volume The target ink droplet velocity The target ink droplet volume determines the stability of the ink droplet's flight trajectory. The line width and thickness of the printed pattern are determined. The ink characteristics are selected based on the two parameters that have the greatest impact on ink droplet formation, including the current ink viscosity and the current ink surface tension. The current ink viscosity affects the ink flowability and jet resistance, while the current ink surface tension affects the integrity of the ink droplet leaving the nozzle. The ink characteristics can be collected in real time and transmitted to the industrial control computer through dedicated detection equipment, such as a rotational viscometer or a pendant drop surface tension meter.

[0047] 202. Obtain iteration parameters, including antennal spacing, antennal attenuation factor, search step size, step size attenuation factor, random direction vector and minimum value to avoid divergence. Initialize the antennal spacing and the search step size to obtain the initial antennal spacing and the initial search step size.

[0048] In this embodiment, the iteration parameters include the core control parameters of the beetle whisker algorithm search process. Specifically, the antennal spacing is used to control the search range, the antennal attenuation factor is used to dynamically shrink the search range, the search step size is used to control the parameter adjustment range, the step size attenuation factor is used to dynamically refine the adjustment range, the random direction vector is used to guide the search direction, and the avoidance of divergence minimum value is used to avoid the extreme case where the denominator in the spray index formula is zero, ensuring the stability of the algorithm calculation. The initialization operation focuses on the antennal spacing and the search step size, assigning reasonable initial values ​​to start the algorithm search, ensuring that the beetle whisker algorithm has a wide global search range when it starts, and gradually focusing on the vicinity of the local optimum in the later stages of iteration, thereby improving the search efficiency.

[0049] 203. Based on the current ink viscosity, 256 voltage points are randomly generated to form an initial waveform;

[0050] In this embodiment, the time interval between the 256 voltage points in the initial waveform is fixed at 10 ns to ensure consistent waveform temporal resolution; the amplitude change rate between adjacent voltage points is ≤5V / This avoids stress concentration in piezoelectric ceramic components due to voltage fluctuations, thus extending their service life. The initial waveform is a multi-pulse structure containing pre-pulse and main pulse to adapt to the basic jetting requirements of most inks. Compared to traditional single-pulse or fixed multi-pulse waveforms, the adjustable freedom of the initial waveform is significantly improved. By adjusting parameters such as pulse interval and amplitude, it adapts to different ink characteristics and optimizes droplet formation, expanding the applicability of this method. Specifically, the combination of the 256 voltage points directly determines the number, interval, and amplitude of the pulses: for high-viscosity inks, in the initial stage and during iterative optimization, the pulse interval is extended by adjusting the voltage point distribution to balance pressure wave attenuation and ensure smooth droplet jetting; for low-viscosity inks, the pulse amplitude distribution is optimized to suppress pressure wave reflection and prevent droplet splitting. During printing, if ink characteristics change or jetting deviation is detected, a new iteration will dynamically adjust these voltage point combinations, thereby adjusting the pulse parameters to achieve adaptive optimization driven by multi-pulse without manual intervention.

[0051] Furthermore, in this embodiment of the invention, the step of outputting the initial waveform to trigger ink droplet ejection and acquiring real-time ink droplet information includes:

[0052] 301. The initial waveform is converted into initial analog drive information and output to the piezoelectric inkjet nozzle through the waveform driver to trigger ink droplet ejection;

[0053] In this embodiment, the initial waveform generated by the industrial control computer is a digital signal (256 voltage data points), which is converted into an analog drive signal (amplitude 0) by the waveform drive control program. The signal (130V) is amplified and conditioned by a waveform driver and then transmitted to the piezoelectric inkjet nozzle via a serial port. After receiving the signal, the piezoelectric ceramic element undergoes an inverse piezoelectric effect and deforms, squeezing the ink chamber to form instantaneous pressure, triggering ink droplet ejection. This ensures precise matching between the drive signal and the piezoelectric inkjet nozzle, solving the ejection abnormality problem caused by signal transmission distortion in the prior art.

[0054] 302. Acquire real-time continuous images of ink droplets using an ink droplet observation instrument;

[0055] In this embodiment, an ink droplet observation instrument is used to capture dynamic images of ink droplets based on the principle of stroboscopic imaging. A nano-pulse stroboscopic light source emits ultrashort pulse light, which is synchronized with the shooting frame rate of the CCD camera to illuminate the area from the nozzle exit to 2mm before landing of the ink droplet. The CCD camera continuously captures and obtains a real-time continuous image sequence of the ink droplet, which completely records the formation, detachment and flight process of the ink droplet, providing high-quality raw data for subsequent feature extraction and solving the defect of insufficient image quality in existing static feedback methods.

[0056] 303. Perform preprocessing, segmentation, and feature extraction processing sequentially on the real-time continuous ink droplet image to obtain real-time ink droplet information;

[0057] In this embodiment, a systematic processing flow is performed on the real-time continuous ink droplet image. Through preprocessing, segmentation, and feature extraction, the image data is transformed into quantified real-time ink droplet information. This addresses the core issues of motion blur, noise, and low contrast in high-speed captured images, significantly improving the quality of the ink droplet image. This provides reliable feedback data support for the optimization calculation of the longhorn beetle whisker algorithm, solves the defects of insufficient ink droplet edge recognition accuracy and inaccurate feature extraction in the prior art, and improves the credibility of the optimization results.

[0058] Further, in this embodiment of the invention, the step of sequentially performing preprocessing, segmentation, and feature extraction processing on the real-time continuous ink droplet image to obtain real-time ink droplet information includes:

[0059] 401. Perform motion blur restoration, noise suppression, and contrast enhancement processing sequentially on the real-time continuous ink droplet image to complete the preprocessing and obtain the preprocessed ink droplet image;

[0060] In this embodiment, to address the issues of motion blur, noise interference, and insufficient contrast in real-time continuous ink droplet images, three processing steps are performed sequentially: motion blur restoration, noise suppression, and contrast enhancement. Specifically, a blind deconvolution algorithm is used to restore the motion blur region, a hybrid filtering algorithm is used to remove ambient light and electronic noise, and an adaptive histogram equalization algorithm is used to enhance the grayscale difference between the ink droplet and the background, ultimately resulting in a high-quality preprocessed ink droplet image.

[0061] 402. Perform adaptive threshold segmentation and edge correction and optimization processing on the preprocessed ink droplet image in sequence to complete the segmentation process and obtain the ink droplet contour positioning image;

[0062] In this embodiment, to achieve accurate separation between the ink droplet region and the background region, a combination of global threshold segmentation, local threshold correction, and edge optimization is adopted. Specifically, firstly, global adaptive threshold segmentation is performed using the Otsu method to initially separate the ink droplet from the background; then, local threshold segmentation (15×15 window) is used to correct the edge region deviation; finally, morphological opening operation and curve fitting algorithm are used to optimize the edge to obtain the ink droplet contour positioning image.

[0063] 403. From the ink droplet contour positioning image, the real-time ink droplet volume is obtained by contour integration and equivalent sphere method, the real-time ink droplet velocity is obtained by inter-frame template matching and sub-pixel positioning algorithm, and auxiliary features of ink droplet morphology are extracted.

[0064] In this embodiment, core features and auxiliary features are extracted from the ink droplet contour localization image. Specifically, the real-time ink droplet volume is calculated using contour integral and equivalent sphere method. Real-time droplet velocity is calculated using inter-frame template matching and sub-pixel localization algorithms. The model extracts morphological auxiliary features such as tail length, number of satellite points, and droplet roundness. The tail length is the straight-line distance between the droplet body and the tip of the tail. The number of satellite points is the number of tiny droplets separated from the droplet body with a diameter ≥ 0.5 μm. The droplet roundness is... ,in, Let the area be the projected area of ​​the ink droplet. The perimeter of the ink droplet.

[0065] 404. Based on the aforementioned ink droplet morphology auxiliary features, the ink droplet category is determined using a regularized threshold judgment algorithm;

[0066] In this embodiment, the ink droplet category is determined based on morphological auxiliary features using the regularized threshold judgment algorithm. Specifically, the types of ink droplets include:

[0067] Category 0, no ink droplets formed: ink droplet projected area < 10 ;

[0068] Class 1, Single droplet: Trail length ≤ 5 Satellite point count = 0; ink droplet roundness ≥ 0.85;

[0069] Category 2, Satellite Droplets: Number of satellite points ≥ 1 or trail length > 5 Furthermore, the roundness of the ink droplets is <0.85;

[0070] The three-part design of the ink droplet categories is well-suited to the jet index formula of this embodiment, enabling the beetle whisker algorithm to clearly identify the ink droplet defect type and then adjust the waveform parameters accordingly, such as optimizing the pulse falling edge slope to suppress tailing and adjusting the pulse interval to suppress satellite points, significantly improving the targeting and efficiency of optimization.

[0071] 405. Integrate the ink droplet type, the real-time ink droplet velocity, and the real-time ink droplet volume to obtain real-time ink droplet information;

[0072] In this embodiment, the integrated real-time ink droplet information is stored in the format of "timestamp + ink droplet type + real-time ink droplet speed + real-time ink droplet volume + morphological auxiliary features", providing multi-dimensional data input for the optimization calculation of the longhorn beetle whisker algorithm.

[0073] Furthermore, in this embodiment of the invention, the step of generating an optimized waveform using the beetle whisker algorithm based on the target parameters, the real-time ink droplet information, and the initialized iteration parameters includes:

[0074] 501. Substitute the target parameters and the real-time ink droplet information into the pre-constructed mapping quality index formula to obtain the current jetting quality index;

[0075] In this embodiment, the target parameters and the real-time ink droplet information are substituted into the pre-constructed mapping quality index formula, specifically:

[0076] ;

[0077] in, Indicates the first The position of the longhorn beetle in the next iteration. To avoid diverging to the minimum value, , For the target ink droplet velocity, For the target ink droplet volume, For ink droplet categories, For real-time droplet speed, Real-time ink droplet volume; This term is used to minimize speed deviation and encourage droplet speed to approach the target droplet speed; This item is used to minimize volume deviation and ensure the accuracy of line width and thickness of the printed pattern; This item is used to optimize the quality of ink droplet formation, thereby guiding the algorithm to generate high-quality waveforms without satellite droplets or trailing. This is a jetting quality indicator used to quantify the quality of ink droplet formation; its value range is [value range missing]. The closer the value is to 1, the better the ink droplet state.

[0078] By pre-constructing a quantitative jetting index formula based on multi-dimensional parameters, discrete parameters such as droplet type, droplet velocity, and droplet volume are integrated into a single continuous quality index. This solves the problems of vague and subjective evaluation standards in existing technologies, providing a clear and quantifiable target for algorithm optimization. Furthermore, the jetting index formula is deeply bound to droplet type, which can specifically suppress printing defects such as satellite droplets, trailing, and droplet splitting, reducing the defect rate of printed patterns and meeting the needs of high-precision scenarios such as OLED display manufacturing and biological cell printing.

[0079] 502. Based on the current spray quality index, determine whether the preset termination search condition is met;

[0080] In this embodiment, the termination search conditions include quality index attainment conditions, iteration count attainment conditions, and optimization gain saturation conditions. The quality index attainment conditions are applicable to scenarios where ink characteristics are stable and printhead condition is good, enabling rapid output of the optimal waveform and maximizing optimization efficiency. Specifically: ≥Preset deviation threshold; The iteration count requirement is used to prevent the algorithm from getting stuck in infinite iteration and to ensure that the optimization process is completed within a specified time. It is applicable to scenarios with large fluctuations in dynamic operating conditions. Specifically, the cumulative iteration count reaches a preset maximum value (50 (100 iterations, default 80 iterations); The optimization gain saturation condition is used to identify the convergence state of the algorithm, avoid unnecessary iterations that consume computing power, and is suitable for scenarios with complex parameter spaces and small fluctuations near the optimal solution. Specifically, if the quality index improvement of 5 consecutive iterations is ≤0.005, it indicates that the algorithm is close to the global optimum, and further iterations will not yield significant gains. The triple termination search condition ensures both optimization accuracy and avoids unnecessary iterations, balancing optimization effect and efficiency. Compared with traditional optimization methods with a single termination condition, the optimization time is shortened.

[0081] 503. If not satisfied, then use the beetle whisker algorithm to generate an optimized waveform based on the initialized iteration parameters; if satisfied, then stop generating the optimized waveform.

[0082] In this embodiment, if the current spray quality index If the termination search condition is not met, it means that there is still room for optimization in the initial waveform. Therefore, based on the initialized iteration parameters, the beetle whisker algorithm is started to adjust the waveform parameters of 256 voltage points to generate the initial optimized waveform. If the termination search condition is met, it means that the current waveform has reached the optimal state. The optimization process is stopped, and the current waveform is output as the final optimized waveform. The beetle whisker algorithm does not require gradient information. It can efficiently search for the optimal solution in the 256-dimensional voltage parameter space by simulating the search behavior of the beetle's antennae. It has low computational complexity and fast convergence speed. It solves the defects of traditional parameter search methods that are prone to getting trapped in local optima and have a large amount of computation, and improves the probability of finding the global optimal solution.

[0083] Furthermore, in this embodiment of the invention, the step of generating the optimized waveform using the beetle whisker algorithm based on the initialized iterative parameters includes:

[0084] 601. Based on the initial waveform and the random direction vector, calculate the waveform of the left antenna and the waveform of the right antenna;

[0085] In this embodiment, based on the initial waveform of the current iteration and random direction vector The waveform of the left antenna is generated through vector operations. And the waveform of the right antenna The calculation formula is:

[0086] ;

[0087] ;

[0088] ;

[0089] in, The interpupillary spacing in the current iteration. The function can generate random numbers. The parameter dimension is indicated; the waveforms of the left and right antennae differ only in the local voltage parameters, corresponding to the two search directions of the beetle whisker algorithm.

[0090] 602. Based on the waveforms of the left antenna and the right antenna, calculate the left fitness value and the right fitness value;

[0091] In this embodiment, the waveform of the left antenna is... And the waveform of the right antenna The waveform driver outputs sequentially to the piezoelectric inkjet nozzle, triggering ink droplet ejection. Corresponding ink droplet images are acquired by an ink droplet observer, and after processing, information on the left and right antennal droplets is obtained. These information are then substituted into a pre-constructed ejection index formula to calculate the left fitness value. And right fitness value The fitness value is positively correlated with the droplet quality.

[0092] 603. Update the initial waveform based on the left fitness value and the right fitness value to obtain the initial optimized waveform;

[0093] In this embodiment, the left fitness value is compared. Right fitness value ,like > Then, adjust the initial waveform along the left antenna direction, and update the formula as follows:

[0094] ;

[0095] like > Then, adjust the initial waveform along the direction of the right antenna, and update the formula as follows:

[0096] ;

[0097] Obtain the initial optimized waveform ,Right now The position of the longhorn beetle in the next iteration; where... It is a symbolic function.

[0098] 604. Update the initial antennal spacing and the initial search step size to obtain the updated antennal spacing and the updated search step size;

[0099] In this embodiment, the antennal spacing and search step size are updated according to a preset attenuation factor to achieve dynamic shrinkage of the search range and adjustment amplitude; specifically, the antennal spacing is updated as follows: Update the search step size to .

[0100] 605. Based on the updated antenna spacing and updated search step size, determine whether the preset iterative search stopping condition is met, and based on the left fitness value and the right fitness value, determine whether the preset termination search condition is met.

[0101] In this embodiment, the stopping condition for the iterative search is: determining the updated antennal spacing. Check if the value is greater than or equal to the minimum value to avoid divergence, where the minimum value to avoid divergence is 0.01, and update the search step size. Whether it is ≥ the minimum step size threshold, where the minimum step size threshold is 0.1V; the termination search condition is the same as in step 502.

[0102] 606. If any condition is not met, the initial waveform is replaced with the initial optimized waveform, and the process returns to the step of calculating the left and right antenna waveforms based on the initial waveform and the random direction vector; if both conditions are met, the optimized waveform is output.

[0103] In this embodiment, the iterative search stopping condition is determined first. If the updated antenna spacing or the updated search step size is lower than the threshold, it indicates that the search range is small enough, and continued iteration may lead to algorithm divergence, so the iteration is terminated directly. Then, the termination search condition is determined. If the fitness value has reached the target, the number of iterations has been reached, or the gain has saturated, the iteration is terminated and the optimal waveform is output. If the iterative search stopping condition or the termination search condition is not met, the algorithm returns to the antenna waveform generation step to start the next round of iteration. The dual condition judgment mechanism ensures the stability and optimization accuracy of the algorithm: the iterative search condition prevents the algorithm from diverging due to excessively small parameters, and the termination search condition ensures that the optimization effect meets the preset requirements. The combination of the two enables the algorithm to search efficiently while avoiding invalid iterations and system oscillations, thus improving the stability of the optimization process.

[0104] Furthermore, in this embodiment of the invention, obtaining a preset deviation threshold and determining whether a state deviation problem exists based on the real-time optimized ink droplet information and the preset deviation threshold includes:

[0105] 701. Obtain the preset deviation threshold;

[0106] In this embodiment, the deviation threshold is the preset optimal jet index threshold, which is preset and stored in the industrial control computer based on the accuracy requirements of the printing scenario. It can be manually adjusted or imported into the preset configuration file through the human-machine interface of the waveform drive control program.

[0107] 702. Input the real-time optimized droplet information and the target parameters into the pre-constructed mapping quality index formula to obtain the optimized jetting quality index;

[0108] In this embodiment, the optimized jet quality index is used to quantify the state of ink droplets based on the optimized waveform.

[0109] 703. Compare the optimized spray quality index with the preset deviation threshold to determine whether there is a state deviation problem;

[0110] In this embodiment, the actual effect of real-time monitoring of the optimized waveform can detect state deviation problems in a short time after dynamic changes in working conditions, and automatically start the iterative correction process to quickly update the optimized waveform, ensuring the continuity and stability of printing quality. This solves the defects of poor universality of optimization results and susceptibility to dynamic interference in the prior art. The alarm mechanism for state deviation provides users with a fault warning function, avoiding large-scale printing defects and material waste caused by abnormal working conditions, and reducing production risks.

[0111] The inkjet waveform optimization method in the embodiments of the present invention has been described above. The inkjet waveform optimization device in the embodiments of the present invention will be described below. Please refer to [link / reference]. Figure 2 One embodiment of the inkjet waveform optimization device in this invention includes:

[0112] The acquisition module 801 is used to acquire target parameters, iteration parameters, and ink characteristics, generate an initial waveform based on the ink characteristics, and initialize the iteration parameters.

[0113] The first output module 802 is used to output the initial waveform to trigger ink droplet ejection and to acquire real-time ink droplet information;

[0114] The first iteration module 803 is used to generate an optimized waveform based on the target parameters, the real-time ink droplet information and the initialized iteration parameters using the beetle whisker algorithm.

[0115] The second output module 804 is used to output the optimized waveform to trigger ink droplet ejection and to obtain real-time optimized ink droplet information.

[0116] The judgment module 805 is used to obtain a preset deviation threshold and determine whether there is a state deviation problem based on the real-time optimized ink droplet information and the preset deviation threshold.

[0117] The second iteration module 806 is used to replace the real-time ink drop information with the real-time optimized ink drop information if there is a state deviation problem, and return to execute the optimization waveform generated by the beetle whisker algorithm based on the target parameters, the real-time ink drop information and the initialized iteration parameters.

[0118] Based on the same ideas as the methods in the above embodiments, the apparatus provided in this application can implement the methods in the above embodiments.

[0119] aboveFigure 2 The inkjet waveform optimization device in this embodiment of the invention will be described in detail from the perspective of modular functional entities. The inkjet waveform optimization device in this embodiment of the invention will be described in detail from the perspective of hardware processing.

[0120] Figure 3 This is a schematic diagram of the structure of an inkjet waveform optimization device 900 provided in an embodiment of the present invention. The inkjet waveform optimization device 900 can vary significantly due to different configurations or performance characteristics. It may include one or more central processing units (CPUs) 910 and a memory 920, and one or more storage media 930 (e.g., one or more mass storage devices) storing application programs 933 or data 932. The memory 920 and storage media 930 can be temporary or persistent storage. The program stored in the storage media 930 may include one or more modules (not shown in the diagram), each module including a series of instruction operations on the inkjet waveform optimization device 900. Furthermore, the processor 910 may be configured to communicate with the storage media 930 and execute the series of instruction operations in the storage media 930 on the inkjet waveform optimization device 900 to implement the steps of the inkjet waveform optimization method provided in the above-described method embodiments.

[0121] The inkjet waveform optimization device 900 may also include one or more power supplies 940, one or more wired or wireless network interfaces 950, one or more input / output interfaces 960, and / or one or more operating systems 931, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will understand that... Figure 3 The illustrated structure of the inkjet waveform optimization device does not constitute a limitation on the inkjet waveform optimization device. It may include more or fewer components than illustrated, or combine certain components, or have different component arrangements.

[0122] 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 inkjet waveform optimization method.

[0123] 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.

[0124] 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.

[0125] 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.

Claims

1. A method for optimizing inkjet waveforms, characterized in that, The inkjet testing system includes an industrial control computer and a piezoelectric inkjet nozzle, a waveform driver, and an ink droplet observer, all electrically connected to the industrial control computer. The industrial control computer outputs an initial waveform or an optimized waveform to the piezoelectric inkjet nozzle through the waveform driver to trigger ink droplet ejection. The ink droplet observer is used to acquire real-time ink droplet information. The inkjet waveform optimization method includes: Obtain the target parameters, iteration parameters, and ink characteristics; generate an initial waveform based on the ink characteristics; and initialize the iteration parameters. The initial waveform is output to trigger ink droplet ejection and real-time ink droplet information is acquired; Based on the target parameters, the real-time ink droplet information, and the initialized iteration parameters, the beetle whisker algorithm is used to generate an optimized waveform. The optimized waveform is output to trigger ink droplet ejection, and real-time optimized ink droplet information is obtained; A preset deviation threshold is obtained, and a state deviation problem is determined based on the real-time optimized ink droplet information and the preset deviation threshold. If there is a state deviation problem, the real-time ink drop information is replaced with the real-time optimized ink drop information, and the process is returned to execute the optimization waveform generated by the beetle whisker algorithm based on the target parameters, the real-time ink drop information and the initialized iteration parameters.

2. The inkjet waveform optimization method according to claim 1, characterized in that, The steps of acquiring target parameters, iteration parameters, and ink characteristics, generating an initial waveform based on the ink characteristics, and initializing the iteration parameters include: Obtain target parameters and ink characteristics, wherein the target parameters include target droplet velocity and target droplet volume, and the ink characteristics include current ink viscosity and current ink surface tension; Obtain iteration parameters, including antennal spacing, antennal attenuation factor, search step size, step size attenuation factor, random direction vector, and minimum value to avoid divergence. Initialize the antennal spacing and the search step size to obtain the initial antennal spacing and the initial search step size. Based on the current ink viscosity, 256 voltage points are randomly generated to form the initial waveform.

3. The inkjet waveform optimization method according to claim 1, characterized in that, The process of outputting the initial waveform to trigger ink droplet ejection and acquiring real-time ink droplet information includes: The initial waveform is converted into initial analog drive information and output to the piezoelectric inkjet nozzle through the waveform driver to trigger ink droplet ejection; Real-time continuous images of ink droplets were acquired using an ink droplet observation instrument; The real-time ink droplet continuous image is subjected to preprocessing, segmentation, and feature extraction processes in sequence to obtain real-time ink droplet information.

4. The inkjet waveform optimization method according to claim 3, characterized in that, The real-time ink droplet continuous image is sequentially preprocessed, segmented, and feature extracted to obtain real-time ink droplet information, including: The real-time continuous ink droplet image is sequentially subjected to motion blur restoration processing, noise suppression processing, and contrast enhancement processing to complete the preprocessing and obtain a preprocessed ink droplet image. The preprocessed ink droplet image is sequentially subjected to adaptive threshold segmentation and edge correction and optimization processing to complete the segmentation process and obtain the ink droplet contour positioning image. From the ink droplet contour localization image, the real-time ink droplet volume is obtained by contour integration and equivalent sphere method, the real-time ink droplet velocity is obtained by inter-frame template matching and sub-pixel localization algorithm, and auxiliary features of ink droplet morphology are extracted. Based on the aforementioned ink droplet morphology auxiliary features, the ink droplet category is determined using a regularized threshold judgment algorithm; By integrating the ink droplet type, the real-time ink droplet velocity, and the real-time ink droplet volume, real-time ink droplet information is obtained.

5. The inkjet waveform optimization method according to claim 2, characterized in that, The step of generating an optimized waveform using the beetle whisker algorithm based on the target parameters, the real-time ink droplet information, and the initialized iteration parameters includes: Substituting the target parameters and the real-time ink droplet information into the pre-constructed mapping quality index formula, the current jetting quality index is obtained; Based on the current injection quality index, determine whether the preset termination search condition is met; If the conditions are not met, the optimized waveform is generated using the beetle whisker algorithm based on the initialized iteration parameters; if the conditions are met, the generation of the optimized waveform is stopped.

6. The inkjet waveform optimization method according to claim 5, characterized in that, The process of generating optimized waveforms using the beetle whisker algorithm based on initialized iterative parameters includes: Based on the initial waveform and the random direction vector, calculate the waveforms of the left and right antennae; Based on the waveforms of the left and right antennae, calculate the left fitness value and the right fitness value; The initial waveform is updated based on the left fitness value and the right fitness value to obtain the initial optimized waveform; Update the initial antennal spacing and the initial search step size to obtain the updated antennal spacing and the updated search step size; Based on the updated antenna spacing and updated search step size, it is determined whether the preset iterative search stopping condition is met, and based on the left fitness value and the right fitness value, it is determined whether the preset termination search condition is met. If any condition is not met, the initial waveform is replaced with the initial optimized waveform, and the process returns to the step of calculating the left and right antenna waveforms based on the initial waveform and the random direction vector; if both conditions are met, the optimized waveform is output.

7. The inkjet waveform optimization method according to claim 1, characterized in that, The step of obtaining a preset deviation threshold and determining whether a state deviation problem exists based on the real-time optimized ink droplet information and the preset deviation threshold includes: Obtain the preset deviation threshold; The real-time optimized droplet information and the target parameters are input into a pre-constructed mapping quality index formula to obtain the optimized jetting quality index. The optimized injection quality index is compared with a preset deviation threshold to determine whether there is a state deviation problem.

8. An inkjet waveform optimization device, characterized in that, include: The acquisition module is used to acquire target parameters, iteration parameters, and ink characteristics, generate an initial waveform based on the ink characteristics, and initialize the iteration parameters. The first output module is used to output the initial waveform to trigger ink droplet ejection and to acquire real-time ink droplet information; The first iteration module is used to generate an optimized waveform based on the target parameters, the real-time ink droplet information, and the initialized iteration parameters, using the beetle whisker algorithm. The second output module is used to output the optimized waveform to trigger ink droplet ejection and to obtain real-time optimized ink droplet information; The judgment module is used to obtain a preset deviation threshold and determine whether there is a state deviation problem based on the real-time optimized ink droplet information and the preset deviation threshold. The second iteration module is used to replace the real-time ink drop information with the real-time optimized ink drop information if there is a state deviation problem, and return to execute the optimization waveform generated by the beetle whisker algorithm based on the target parameters, the real-time ink drop information and the initialized iteration parameters.

9. An inkjet waveform optimization device, characterized in that, The inkjet waveform optimization device includes: a memory and at least one processor, wherein the memory stores instructions; At least one of the processors invokes the instructions in the memory to cause the inkjet waveform optimization device to perform the steps of the inkjet waveform optimization method as claimed in any one of claims 1-7.

10. A computer-readable storage medium storing instructions thereon, characterized in that, When the instructions are executed by the processor, they implement the steps of the inkjet waveform optimization method as described in any one of claims 1-7.

Images