Local cleaning of print heads in printing systems

Abstract

Apparatus, method, and system for localized cleaning of printheads in a printing system. [Solution] The printing system performs a printhead health check operation on the printhead of the printing system. The printhead health check operation includes identifying and recording a set of nozzle defects in the nozzles of the printhead. Nozzle defects may include any of the following: clogs, misfires, and mismatches. Based on the printhead health check operation, the printing system determines a set of nozzle degradation metric values ​​for the printhead to indicate the degree and nature of the nozzle defects. Using the set of nozzle degradation metric values, the printing system generates a set of actions for the carrier associated with the printhead. The printing system performs printhead cleaning according to the generated actions.

Description

Technical Field

[0001] The embodiments described herein generally relate to printing systems. More specifically, the embodiments relate to methods for cleaning and restoring systems of inkjet printheads.

Background Art

[0002] A printhead cleaning system is a subsystem of an industrial inkjet printer. The nozzles of the printhead may be partially or completely clogged due to multiple causes, which may result in areas that are not printed. Causes of printing defects may include air bubbles trapped inside the printhead, foreign matter in the nozzle plate, wetting of the nozzle plate, etc.

[0003] In this specification, systems and methods for local cleaning of the printhead of a printing system are introduced. In some embodiments, the printing system performs a health check operation on the printhead of the printing system. The health check operation of the printhead includes identifying and recording a series of nozzle defects such as clogging, non-firing, and inconsistency in the nozzles of the printhead. Based on the health check operation of the printhead, the printing system determines a set of nozzle degradation metric values indicating the degree and nature of the nozzle defects. The nozzle degradation metric values are associated with the health state of the printhead nozzles. Using the set of nozzle degradation metric values, the printing system generates a set of actions for the carrier associated with the printhead. The carrier can be a movable carriage fixed to the base of the printing system and equipped with a cleaning device configured to clean the printhead. The set of actions adjusts at least one parameter such as the speed of the cleaning device, the movement profile, the vacuum level, and the wiper pressure, and defines the operating strategy of the printhead. The printing system executes the cleaning of the printhead according to the generated actions.

[0004] Exemplary embodiments of this specification include performing a health check operation of the printhead of a printing system by evaluating a printed template using one or more camera sensors of the printing system. Based on this evaluation, the printing system identifies the cleanliness level of the printhead using a carrier camera pointed at the plate surface of the printer nozzle. In addition, the printing system can detect nozzle clogs using a carrier laser emitter-receiver pair and identify interruptions in the expected trajectory of droplets ejected from the nozzle. Abnormal ejection can be evaluated using a carrier nozzle actuator. Identification and recording of nozzle defects includes recording the frequency, severity, and / or location of defects within the printhead nozzle.

[0005] These aspects, features, and embodiments can be expressed as methods, apparatus, systems, components, program products, means or steps for performing functions, and in other ways. These aspects, features, and embodiments will become apparent from the following description, including the claims. [Brief explanation of the drawing]

[0006] [Figure 1] This block diagram shows a perspective view of a printing system according to one or more embodiments. [Figure 2] This is a block diagram showing a side view of a printing system including a printer head and a light source, according to one or more embodiments. [Figure 3] This is a schematic diagram of an exemplary environment for a printhead array and a movable carrier system to perform a cleaning action, according to an embodiment. [Figure 4] This is a schematic diagram of a front view of a single-pass inkjet printer according to an embodiment. [Figure 5] The image shows a processed print image used to evaluate the health of the printhead nozzles using human or computer vision technology, according to an embodiment. [Figure 6]This graph shows an example of the relationship between nozzle degradation metric value and cleaning system speed according to the embodiment. [Figure 7] This is a flowchart for performing a cleaning action according to one embodiment. [Figure 8] This is a high-level block diagram showing an example of an AI system according to one or more embodiments. [Figure 9] This block diagram shows an example of a computer system capable of performing at least some of the operations described herein. [Modes for carrying out the invention]

[0007] In the following description, for the sake of clarity and to fully understand this embodiment, numerous specific details are provided. However, it will be clear that this embodiment can be implemented even without these specific details.

[0008] This specification presents a system, method, and apparatus for localized cleaning of printheads in a printing system. Conventional methods for cleaning printheads in a printing system are typically uniform cleaning processes that apply the same cleaning intensity procedure and / or cleaning time across the entire printhead, regardless of the level of nozzle degradation. As a result, some nozzles may be over-cleaned, leading to unnecessary wear and damage, which can shorten their lifespan and / or increase maintenance costs. Conversely, other nozzles may remain insufficiently cleaned, potentially resulting in persistent print quality problems such as streaks, smudges, or missing lines in the print output. In addition, conventional cleaning methods are time-consuming and resource-intensive because they do not dynamically adapt to changes in the level of nozzle degradation across a particular printhead. Conventional methods not only waste time but also consume excessive amounts of cleaning fluid and energy, so this inefficiency consumes many resources, further increasing operating costs and reducing the overall efficiency of the printing system.

[0009] Embodiments disclosed herein provide a method for performing a health check of a printhead, identifying and recording specific nozzle defects, and determining a set of metric values ​​for nozzle degradation. Based on the set of metric values ​​for nozzle degradation that quantify the degree and nature of the defects, the printing system generates and executes the appropriate cleaning actions. These actions may include adjusting various parameters such as carrier speed, travel profile, vacuum level, and wiper pressure to ensure efficient and effective cleaning. For example, the carrier speed may be reduced in areas with a high concentration of defects for more thorough cleaning, while the vacuum level and wiper pressure may be increased to remove stubborn clogs.

[0010] The advantages and benefits of using a localized approach to clean printheads using the embodiments described herein include improved print quality by ensuring that all nozzles are properly cleaned. By addressing nozzle defects locally, the printing system can produce high-quality prints with consistent color and sharpness, free from streaks and missing lines. In addition, the localized approach extends printhead life by preventing unnecessary wear on nozzles in good condition. By avoiding over-cleaning, the printing system reduces the risk of nozzle damage, thereby lowering maintenance costs and extending the printhead replacement interval. Furthermore, the localized approach reduces downtime and resource consumption, improving the overall productivity and cost-effectiveness of the printing system. By dynamically adjusting cleaning actions based on real-time data, the printing system reduces maintenance time and resources, resulting in more efficient use of the printing system and reduced operating costs. Printing system

[0011] Figure 1 is a block diagram showing a perspective view of a printing system 100 according to one or more embodiments. The printing system 100 includes a printer head 106, at least one light source 112, and a transfer belt 102. Depending on the embodiment, other components such as a dryer may also be included. For example, the light source 112 is present in some embodiments but not in others. In some examples, the printing system 100 in Figure 1 includes a transfer belt 102, but other means may be used to transport and / or hold the workpiece 104.

[0012] The printer head 106 is configured to deposit ink onto the substrate 104 in the form of an image 110. In some embodiments, the printer head 106 is an inkjet printer head that, for example, uses piezoelectric nozzles to spray ink onto the substrate 104. In some embodiments, the ink is solid energy (e.g., UV-curing ink). However, other inks such as aqueous energy-curing inks and solvent-based energy-curing inks are also used.

[0013] In some embodiments, one or more light sources 112 cure some or all of the ink applied to the substrate 104 by emitting UV radiation. In some examples, the light sources 112 are any combination of UV fluorescent bulbs, UV light-emitting diodes (LEDs), low-pressure (e.g., mercury (Hg)) bulbs, or excitation dimer (excimer) lamps and / or lasers. Various combinations of these light sources may be used. In some examples, the printing system 100 includes a low-pressure Hg lamp and a UV LED. The light sources 112 may be configured to emit a specific subtype of UV radiation.

[0014] The printer head 106 and the light source 112 are shown directly adjacent to each other, that is, adjacent without any intervening components. However, in other embodiments, additional components may also be present to assist with printing, curing, etc. In some examples, multiple separate light sources 112 are located behind the printer head 106. Figure 1 illustrates one possible order in which components are arranged to print an image 110 onto a substrate 104. Other embodiments are also conceivable in which additional components are arranged before, between, or after the illustrated components, etc.

[0015] In some embodiments, one or more of the aforementioned components are housed in one or more stations. For example, the printer head 106 may be housed in the printing station 108, and the light source 112 may be housed in the irradiation station 114. In addition to protecting the components from damage, these stations also provide other advantages in some examples. For example, the curing irradiation station 114 limits the portion of the substrate 104 and image 110 that should be exposed during the curing process. The stations may be permanently mounted on the track or chassis of the printing system 100. The substrate 104 moves relative to the printer head 106, light source 112, etc., so that ink adheres to the substrate 104.

[0016] In various embodiments, some or all of the components are controlled by a computer system 116. In some examples, the computer system 116 allows the user to input print instructions and information (e.g., by changing the curing settings) to modify the print settings and the print process.

[0017] Figure 2 is a block diagram showing a side view of a printing system 200, including a printer head 202 and a light source 204, according to one or more embodiments. In some examples, the printer head 202 has different ink / color stages, such as cyan, magenta, yellow, and black (CMYK), which are applied to the surface of the substrate 206. Path A represents the media feeding direction (e.g., the direction in which the substrate 206 moves during the printing process). Path D represents the distance between the printer head 202 and the surface of the substrate 206.

[0018] In some embodiments, the light source 204 cures some or all of the ink 208 applied to the substrate 206 by the printer head 202. In some examples, the light source 204 is configured to emit UV electromagnetic radiation of wavelengths of subtype V (UVV), subtype A (UVA), subtype B (UVB), subtype C (UVC), or any combination thereof. Generally, the wavelength of UVV is measured between 395 nanometers (nm) and 445 nm, the wavelength of UVA is measured between 315 nm and 395 nm, the wavelength of UVB is measured between 280 nm and 315 nm, and the wavelength of UVC is measured between 100 nm and 280 nm. However, those skilled in the art will recognize that these ranges are somewhat adjustable. For example, in some embodiments, a wavelength of 285 nm is characterized as UVC.

[0019] In some examples, the light source 204 is, for example, a fluorescent bulb, a light-emitting diode (LED), a low-pressure (e.g., mercury (Hg)) bulb, or an excitation dimer (excimer) lamp / laser. In some embodiments, different combinations of light sources can be used. Generally, the light source 204 is selected so that the curing temperature does not exceed the temperature at which the ink 208 begins to sublimate. For example, the light source may be a UV LED lamp, which generates little heat and is used for a wide variety of curing materials. UV LED lamps are characterized by low power consumption, long lifespan, and predictable output.

[0020] Alternatively, or in addition, other curing processes such as epoxy (resin) chemistry, flash curing, electron beam technology, etc. are also used. Those skilled in the art will understand that many different curing processes can be employed that utilize specific time frames, strengths, speeds, etc. In some embodiments, the strength increases or decreases linearly or non-linearly (e.g., exponentially, logarithmically). In some embodiments, the strength is changed using a variable resistor or, in the case of an LED light source, by applying a pulse width modulation (PWM) signal to the diode. In some examples, the light is modulated using amplitude modulation, polarization modulation, frequency modulation (e.g., in the case of wavelength division multiplexing (WDM)), phase modulation (e.g., angular phase control), time modulation, etc.

[0021] Some embodiments can be understood by referring to FIG. 3. FIG. 3 is a schematic diagram of an exemplary environment of a print head array and a movable carrier system for performing a cleaning action according to an embodiment. The components of the environment 300 shown in FIG. 3 include a print head array 302, a cleaning carrier 304, a cleaning system 306, system elements 308, a movement actuator 310, a cleaning actuator 312, and a system controller 314. Additional components of the environment 300 include the components of the computer system 900 illustrated and described in more detail by referring to FIG. 9, and / or the components of the printing system 100 illustrated and described in more detail by referring to FIG. 1. Similarly, other embodiments include different components and / or additional components or are connected in different ways.

[0022] The printhead array 302 includes a plurality of printheads (PH1, PH2, PH3, etc.) arranged in a specific configuration to facilitate the printing operation. Each printhead within the array is responsible for discharging ink or other printing materials onto the printing medium. The printhead array 302 is designed to achieve high-resolution printing by controlling the placement of ink droplets. Each printhead includes a set of nozzles that can eject fine ink droplets at high speed, enabling detailed and accurate printing. In some embodiments, the printhead array 302 can be configured in various shapes (linear, staggered, etc.) according to the printing requirements. The printhead array 302 can also include different types of printheads, such as thermal printheads, piezo printheads, continuous inkjet printheads, etc. Thermal printheads utilize heat to generate bubbles and push the ink out of the nozzles. On the other hand, piezo printheads use piezoelectric crystals that change shape when an electric field is applied to eject the ink. Continuous inkjet printheads, on the other hand, continuously eject ink droplets, and the ink droplets are directed towards the printing substrate or deflected into a recycling system.

[0023] The cleaning carrier 304 is a movable platform that carries the cleaning system across the printhead array 302. The cleaning carrier 304 moves along the printhead array 302 to perform nozzle cleaning actions, monitor performance, and / or prevent clogging. The cleaning carrier 304 moves along the length of the printhead array 302, cleaning each nozzle in the set of nozzles. The cleaning carrier 304 can move linearly or along a specific path directed by the system controller 314. In some embodiments, the cleaning carrier 304 can be designed to include different movement mechanisms such as linear actuators, robotic arms, or conveyor belts. Linear actuators control the movement of the cleaning carrier 304, while robotic arms provide flexibility to reach various parts of the printhead array 302. Conveyor belts can be used for continuous movement across the entire printhead array 302. The cleaning carrier 304 can be equipped with sensors to detect its position relative to the printheads.

[0024] The cleaning system 306 comprises various elements and mechanisms for cleaning the printhead nozzles. The cleaning system 306 performs cleaning actions such as wiping, suction, and / or application of cleaning fluid to remove foreign matter and maintain nozzle health. The cleaning system 306 can be used to prevent clogging and ensure consistent print quality. The cleaning system 306 may include a combination of mechanical and fluid cleaning methods to remove ink residue and other contaminants from the nozzles. In some embodiments, the cleaning system 306 includes an ultrasonic cleaner, a blower, or a special cleaning fluid. An ultrasonic cleaner can use high-frequency sound waves to generate cavitation bubbles in the cleaning fluid, thereby removing particles from the nozzles. A blower uses compressed air to blow away foreign matter, while a special cleaning fluid dissolves ink residue. In addition, the printhead and ink supply system can be operated to assist the cleaning actions. Possible actions include ticking (activating the nozzles without ejecting ink droplets), weeping (reducing the vacuum level to cause ink droplets to drip from the printhead), spitting (ejecting ink droplets from the nozzles), and / or reversing the direction of ink flow across the entire printhead. The cleaning system 306 can be integrated with one or more automated diagnostic tools to evaluate the effectiveness of the cleaning process.

[0025] System elements 308 are individual components within the cleaning system 306 that perform specific cleaning tasks. System elements 308 include brushes, wipers, suction heads, and fluid applicators that directly interact with the printhead nozzles. Each element can address different types of contaminants and ensure comprehensive cleaning. Brushes physically clean the nozzle plates to remove dried ink. Wipers sweep and clean the nozzles to remove loose particles. Suction heads create a vacuum to suck foreign matter from the nozzles, and fluid applicators discharge cleaning fluid to dissolve and wash away ink residue. In some embodiments, system elements 308 can be made modular to allow for replacement or customization depending on the type of printhead and the nature of the contaminants. System elements 308 can be equipped with sensors to monitor the status and performance of system elements 308.

[0026] The moving actuator 310 is a device that controls the movement of the cleaning carrier 304 along the print head array 302. The moving actuator 310 enables the cleaning carrier 304 to target specific areas of the print head array 302 for cleaning. The moving actuator 310 can be configured to move the cleaning carrier 304 at various speeds to follow linear or nonlinear paths. In some embodiments, the moving actuator 310 can be an electric motor, a pneumatic cylinder, or a hydraulic system. The moving actuator 310 can be integrated with a feedback system such as an encoder or linear scale to precisely position the cleaning carrier 304 so that it moves as intended by the cleaning process.

[0027] The cleaning actuator 312 controls the action of the cleaning system element 308. The cleaning actuator 312 enables the cleaning system element 308 to perform specific cleaning tasks, such as moving a brush or operating a suction head. The cleaning actuator 312 can be a solenoid (e.g., for moving a wiper or brush), a stepping motor (e.g., for rotation or linear movement), or a piezoelectric actuator (e.g., for delicate cleaning tasks). The cleaning actuator 312 can also perform coordinated cleaning actions in synchronization with the moving actuator 310.

[0028] The system controller 314 is a central processing unit that manages the overall operation of the cleaning system 306. The system controller 314 coordinates the actions of the moving actuator 310 and the cleaning actuator 312, processes sensor data, and executes cleaning algorithms. The system controller 314 can coordinate the actions of various components as needed. In some embodiments, the system controller 314 can be integrated with machine learning algorithms to improve the cleaning strategy based on real-time data. Examples of machine learning algorithms are illustrated with reference to Figures 7 and 8.

[0029] Several embodiments can be understood by referring to Figure 4. Figure 4 is a schematic diagram showing a front view of a single-pass inkjet printer 400 according to one embodiment. Figure 4 illustrates various methods for evaluating print quality or printhead nozzle health. The components of the single-pass inkjet printer 400 shown in Figure 4 include a printer bar 402, a receiver pair 404, a camera 406, a substrate 408, and / or a scanner 410. Additional components of the single-pass inkjet printer 400 include components of a computer system 900, which is illustrated and described in more detail with reference to Figure 9, and / or components of a printing system 100, which is illustrated and described in more detail with reference to Figure 1. Similarly, other embodiments include different components and / or additional components, or are connected in different ways.

[0030] The printer bar 402 is a structural component that holds one or more print heads in a fixed position relative to the substrate 408. The printer bar 402 ensures that the print heads are correctly aligned for printing operations and provides a stable platform for the print heads. The printer bar 402 can be made of a rigid material to maintain alignment and prevent movement that may affect print quality. In some embodiments, the printer bar 402 is adjustable to allow for fine-tuning of the print head position. In addition, the printer bar 402 can be integrated with sensors that continuously monitor the print head alignment to correct deviations and maintain print quality.

[0031] In some embodiments, a laser emitter-receiver pair 404 is used for print quality or printhead nozzle health assessment. The laser emitter and corresponding receiver work together to detect anomalies in ink ejection from the printhead. The laser emitter irradiates a beam across the inkjet path, and the receiver detects interruptions or deviations in the beam to indicate potential problems with the printhead nozzle. In some embodiments, other types of light sources, such as LEDs, are used instead of lasers. In some embodiments, multiple emitter-receiver pairs are used to cover a wider area or to improve detection accuracy.

[0032] In some embodiments, a camera 406 is used to visually inspect the surface of the printhead nozzle for print quality or printhead nozzle health assessment. The camera 406 captures high-resolution images of the nozzle surface to identify blockages, debris, or other defects that may affect print quality. The camera 406 can be positioned so that the nozzle plate is clearly visible, enabling the capture of images that can be analyzed for signs of wear or contamination. In some embodiments, a multi-camera system with multiple cameras 406 can be used to provide different fields of view.

[0033] The substrate 408 is the material on which the ink is printed. The substrate 408 can be paper, plastic, fabric, or other printable surface. The substrate 408 acts as the medium for the final printed product and interacts with the print head during the printing process. Different materials of the substrate may have different properties that affect ink adhesion or drying time. In some embodiments, different types of substrates can be used depending on the application, such as glossy paper for high-resolution images or woven fabric for fabric printing. The substrate 408 can be pre-treated to improve ink adhesion and print quality, resulting in sharper and more vivid printed images.

[0034] In some embodiments, print quality or printhead nozzle health assessment uses a scanner 410 and computer vision technology to evaluate the quality of the print output. The scanner 410 scans the printed substrate 408 to detect defects such as misalignment, color inconsistencies, and ink depletion. The data collected by the scanner 410 is used to evaluate the performance of the printhead and make any necessary adjustments. In some embodiments, the scanner 410 is equipped with spectral analysis capabilities for color accuracy. Spectral analysis provides detailed information about the color characteristics of the print output to ensure color accuracy and consistency.

[0035] Several embodiments can be understood by referring to Figure 5. Figure 5 shows a print image 500 processed to evaluate the health of the print head nozzles using human or computer vision technology according to one embodiment. Components of the print image 500 shown in Figure 5 include nozzle defects 502, 504.

[0036] Nozzle defects 502 relate to any problems with the printhead nozzles that may affect print quality. Common defects include nozzle clogging, misalignment, or damage. Nozzle defects 502 can lead to reduced print quality, such as streaks, missing lines, or uneven color. In some embodiments, nozzle defects 502 can be detected using various methods, such as visual inspection, acoustic sensing, and electrical impedance measurement. Various visual inspections involve taking high-resolution images of the nozzle surface and analyzing signs of clogging or damage. Acoustic sensing uses sound waves to detect nozzle clogging or irregularities, while electrical impedance measurement can identify changes in the electrical characteristics of a defective nozzle. In some embodiments, nozzle defects 502 may include partial clogging, where the nozzle is not completely clogged but is still restricted, which leads to reduced ink flow and print quality.

[0037] Several embodiments can be understood by referring to Figure 6. Figure 6 is a graph 600 showing a possible relationship between nozzle degradation metric value and cleaning system speed according to an embodiment. The components of graph 600 shown in Figure 6 include the nozzle degradation metric 602, cleaning carrier speed 604, nozzle degradation metric value 606, and cleaning carrier speed metric value 608. Additional components of graph 600 include components of computer system 900, which are illustrated and described in more detail with reference to Figure 9. Similarly, other embodiments include different components and / or additional components, or are connected in different ways.

[0038] The nozzle degradation metric 602 is a quantitative measure of the health or performance of the nozzles within the printhead. The nozzle degradation metric 602 indicates the degree of nozzle clogging, non-ejection, or other degradation and is used to determine the appropriate cleaning action required to restore the nozzles to optimal condition. The nozzle degradation metric 602 can be derived from various sources, such as optical scanning, jet performance analysis, and electrical measurements. In some embodiments, optical scanning involves visually inspecting the nozzles using a camera or scanner to detect clogging or misalignment. Jet performance analysis monitors the ink ejection pattern and volume to identify discrepancies, while electrical measurements evaluate the electrical properties of the nozzles to detect anomalies.

[0039] The cleaning carrier speed 604 is the speed at which the cleaning mechanism (i.e., the carrier) moves across the printhead to perform the cleaning action. The cleaning carrier speed ensures effective cleaning while reducing time and resource usage. The cleaning carrier speed 604 can be adjusted to improve the efficiency of the cleaning process based on a nozzle degradation metric. In some embodiments, the speed can be fixed, i.e., kept constant regardless of the nozzle condition. In some embodiments, the speed can be made variable by dynamically adjusting it based on a nozzle degradation metric to provide stronger cleaning in heavily degraded areas and faster cleaning in less affected areas, where the direction of movement is constant.

[0040] The nozzle degradation metric value 606 represents a specific measurement or level of nozzle degradation. In Figure 6, the nozzle degradation metric value 606 is plotted on a graph to show the degree of degradation at different locations along the printhead. The nozzle degradation metric value 606 provides a map of nozzle conditions, enabling targeted cleaning actions. In some embodiments, the nozzle degradation metric value 606 can be expressed in various forms such as a percentage, a numerical score, or a categorical classification. A percentage indicates the proportion of affected nozzles, a numerical score provides a quantitative measure based on a predefined scale, and a categorical classification groups degradation into categories such as low, medium, and high.

[0041] The cleaning carrier speed metric value 608 indicates the speed of the cleaning carrier at different positions along the printhead, corresponding to the nozzle degradation metric value. The cleaning carrier speed metric value 608 ensures that the cleaning carrier moves at an appropriate speed based on the nozzle condition to effectively clean the nozzle. In some embodiments, the cleaning carrier speed metric value 608 can be expressed in various forms such as a percentage, absolute speed units, or adaptive speed adjustment. Percentages represent the speed relative to the maximum speed, absolute speed units provide specific measurements in units such as mm / s or inches / s, and adaptive speed adjustment allows for real-time changes based on feedback from sensors.

[0042] Several embodiments can be understood by referring to Figure 7, which is a flowchart for performing a cleaning operation according to an embodiment. In some embodiments, the process in Figure 7 is performed using a printing system, for example, the printing system of environment 300, which is illustrated and described in more detail with reference to Figure 3. In other embodiments, the process in Figure 7 is performed using components of computer system 900, which is illustrated and described in more detail with reference to Figure 9. Similarly, other embodiments include different steps and / or additional steps, or are performed in a different order.

[0043] In step 702, the printing system performs a printhead health check operation on the printhead of the printing system. The printhead health check operation includes identifying and recording a set of nozzle defects in the set of printer nozzles located within the printhead. Nozzle defects in a set of nozzle defects include clogs, misfires, and / or mismatches in the set of printer nozzles within the printhead. The health check can be performed using a variety of methods. Identification and recording of nozzle defects includes identifying and recording the frequency, severity, and / or location of nozzle defects within the printer nozzles.

[0044] In some embodiments, a printhead health check operation includes evaluating a printed template using one or more camera sensors in the printing system. In some embodiments, the printhead health check operation includes capturing and analyzing a printed test pattern using an optical sensor or imaging device associated with the printing system. The optical sensor or imaging device can be configured to detect deviations or defects in the print output printed using the printhead. Based on this evaluation, the printing system can identify the cleanliness level of the printhead using a carrier camera pointed at the plate surface of one or more printer nozzles in a set of printer nozzles. For example, a high-resolution camera can capture an image of the printed test pattern, and image processing software can analyze the image to detect missing or misaligned dots indicating nozzle defects.

[0045] In some embodiments, a printhead health check operation includes detecting nozzle failure in a set of nozzles using a carrier laser emitter-receiver pair in response to the beam of the laser emitter-receiver pair crossing the expected trajectory of an ink droplet ejected from one or more printer nozzles in the printhead's set of printer nozzles. The printing system can indicate a nozzle that is not ejecting properly by projecting a laser beam across the path of the ink droplet and detecting an interruption in the beam. For example, if the laser beam is interrupted, this suggests that the ink droplet is not being ejected properly and indicates a possible clog or mis-ejection.

[0046] In some embodiments, printhead health check operations include evaluating abnormal ejection using the carrier nozzle actuator. For example, a printing system can monitor the electrical signals and mechanical movements of the nozzle actuator to detect abnormalities in these operations. For example, a piezoelectric sensor can measure the vibration of the nozzle actuator to identify deviations from normal operation that may indicate a defect.

[0047] Step 704 determines a set of nozzle degradation metric values ​​for the printhead of the printing system, based on the printhead health check operation, which indicates the set of nozzle defects in the set of printer nozzles. The set of nozzle degradation metric values ​​is associated with the health status of the printhead nozzles. The set of nozzle degradation metric values ​​can indicate, for example, the number of defective nozzles per unit area, the average severity of defects across the entire printhead, and / or the distribution pattern of nozzle defects within the printhead. The nozzle degradation metric values ​​provide a quantitative measure of the printhead's condition. For example, if defects are concentrated in a particular area, it can indicate that that area requires more intensive cleaning. The system uses the set of nozzle degradation metric values ​​to prioritize cleaning actions and effectively allocate resources so that the necessary attention goes to the most critical areas. For example, if the degradation metric indicates that 20% of the nozzles in a particular section are clogged, the system can focus its cleaning efforts on that section.

[0048] In step 706, the printing system generates a set of actions for the carrier associated with the printhead of the printing system, based on a set of nozzle degradation metric values. The carrier is a movable carriage fixed to the base of the printing system and equipped with a cleaning device configured to clean the printhead. The set of actions adjusts at least one parameter of the carrier. The movement of the carrier can be controlled by an actuator and an actuator controller.

[0049] To translate nozzle degradation metrics into actionable cleaning steps, the printing system can use predetermined mapped actions or one or more models as described with reference to Figure 8. For example, the system collects data on nozzle performance, including defect frequency, severity, and location, as well as the effectiveness of various cleaning actions. The data is input into an artificial intelligence (e.g., machine learning) model trained to recognize patterns and correlations between nozzle degradation metrics and effective cleaning strategies. The machine learning model used can be a supervised learning model such as a neural network, decision tree, or support vector machine. During the training phase, the machine learning model is provided with labeled training data. Input features include nozzle degradation metrics (e.g., number of defective nozzles per unit area, average defect severity, defect distribution pattern) and corresponding cleaning actions previously applied (e.g., carrier velocity, movement profile, vacuum level, wiper pressure). Output labels are the effectiveness of the cleaning action, measured by the improvement in nozzle performance after cleaning. The model learns to map input features to output labels to identify one or more cleaning actions for different degradation scenarios.

[0050] Once training is complete, the ML model can build a set of behaviors for any nozzle degradation metric. When a new health check is performed, the system inputs the degradation metric into the model. Input features to the model include current nozzle degradation metrics such as the number of defective nozzles per unit area, the average severity of defects, and the distribution pattern of defects within the printhead. The model processes these inputs using learned parameters and outputs a custom cleaning plan. The model's output can include specific adjustments to carrier speed, travel profile, vacuum level, and wiper pressure. For example, the model might output recommendations to reduce the carrier speed to 0.5 cm / s, increase the vacuum level to -80 kPa, and increase the contact pressure with the wiper in areas with high clogging.

[0051] In some embodiments, the printing system continuously updates its model with new data from each cleaning cycle, allowing the model to refine its predictions and improve cleaning efficiency over time. For example, the printing system can collect post-cleaning performance data, such as reduced nozzle failures and improved print quality, and feed this data back into the model for further training. For example, if the printing system detects that a particular cleaning action was highly effective in clearing clogs in a particular area... If The printing system would assign a high weight to that action in similar future scenarios. Conversely, if the cleaning action was not very effective, the model would adjust its parameters so as not to recommend that action in similar situations.

[0052] A set of actions may include adjusting the carrier speed based on the density of nozzle defects within a specific location on the printhead. For example, the carrier may move more slowly in areas with a high concentration of defects to ensure a more thorough cleaning. In some embodiments, a set of actions may include modifying the carrier's movement profile based on the density of nozzle defects within a specific location on the printhead. This may include modifying the carrier's movement path or pattern to ensure that all affected areas are properly cleaned. For example, the carrier may follow a zigzag pattern in heavily contaminated areas to ensure multiple passes. In some embodiments, a set of actions may include changing the vacuum level of a vacuum pad equipped on the carrier or the contact pressure of a wiper equipped on the carrier based on the density of nozzle defects at a particular location. In areas with more severe contamination, a higher vacuum level or higher contact pressure may be required to ensure effective cleaning. For example, the system may use a vacuum level of -40kPa when the clogging is minor, while increasing the vacuum level to -80kPa in certain areas when the clogging is severe.

[0053] In step 708, the printing system moves the carrier on the print head according to a set of actions. At least one parameter of the carrier to be adjusted includes adjusting the speed at which the carrier moves on the print head. The carrier speed can be adjusted to increase in response to a higher density of nozzle defects in a particular location. This allows the carrier to spend more time cleaning more severely problematic areas. For example, the carrier speed can be reduced to 0.5 cm / s in heavily contaminated areas and increased to 2 cm / s in cleaner areas. The parameter to be adjusted includes the carrier movement profile based on the distribution pattern of nozzle defects within the print head. By changing the carrier movement profile based on the distribution pattern of nozzle defects, targeted cleaning is possible, thereby ensuring that all affected areas are treated. In some embodiments, the parameter to be adjusted includes the vacuum level of the cleaning device. By adjusting the vacuum level of the cleaning device to increase in response to a higher density of nozzle defects, sufficient cleaning action is ensured to remove stubborn contaminants. The vacuum level of the carrier can be adjusted to increase in response to a higher density of nozzle defects in a particular location. In some embodiments, the parameters to be adjusted include the contact pressure of a wiper mounted on the carrier. By adjusting the wiper contact pressure to increase in response to a high density of nozzle defects, it is ensured that the wiper can effectively remove debris from the nozzle plate surface.

[0054] In some embodiments, moving a carrier with at least one adjusted parameter across the printhead includes using a wiper to remove material from the plate surface of one or more printer nozzles in a set of printer nozzles. The printing system may apply a vacuum pad to detach particles from the plate surface of one or more printer nozzles. The printing system may spray a flushing fluid onto the plate surface to remove one or more nozzle defects. For example, a high-pressure flushing fluid may be used to remove stubborn clogs, while a vacuum pad can remove the removed particles. Exemplary Embodiment of a Model Used for Local Cleaning of Printheads

[0055] Figure 8 is a block diagram showing an exemplary artificial intelligence (AI) system 800 according to one or more embodiments of the present disclosure. The AI ​​system 800 is implemented using components of an exemplary computer system 900, which is illustrated and described in more detail with reference to Figure 9. For example, the AI ​​system 800 can be implemented using a processor 902 and memory 906 with programmed instructions 908, which are illustrated and described in more detail with reference to Figure 9. Similarly, implementations of the AI ​​system 800 may include different components and / or additional components, or may be connected in different ways.

[0056] As illustrated, the AI ​​system 800 may include a set of layers in which the architecture of the AI ​​system conceptually organizes elements in an exemplary network topology for implementing a particular AI model 830. Generally, the AI ​​model 830 is a computer executable program implemented by the AI ​​system 800 that analyzes data and makes predictions. Information can pass through each layer of the AI ​​system 800 to produce the output of the AI ​​model 830. These layers include the data layer 802, the structure layer 804, the model layer 806, and the application layer 808. The algorithm 816 of the structure layer 804 and the model structure 820 and model parameters 822 of the model layer 806 together form the exemplary AI model 830. The optimizer 826, the loss function engine 824, and the regularization engine 828 function to refine and optimize the AI ​​model 830, and the data layer 802 provides resources and support for the application of the AI ​​model 830 by the application layer 808.

[0057] The data layer 802 serves as the foundation for the AI ​​system 800 by preparing the data for the AI ​​model 830. As shown in the figure, the data layer 802 comprises two sub-layers: a hardware platform 810 and one or more software libraries 812. The hardware platform 810 can be designed to perform operations on the AI ​​model 830 and may include computing resources for storage, memory, logic, and networking, as described in relation to Figure 9. The hardware platform 810 can process large amounts of data using one or more servers. The servers can perform backend operations such as matrix calculations, parallel computing, and machine learning (ML) training. Examples of servers used in the hardware platform 810 include a central processing unit (CPU) and a graphics processing unit (GPU). A CPU is an electronic circuit designed to execute computer program instructions such as arithmetic, logic, control, and input / output (I / O) operations, and can be implemented on an integrated circuit (IC) microprocessor. A GPU is an electronic circuit originally designed for graphics operation and output, but can also be used for AI applications because it has enormous computing and memory resources. Because GPUs use a parallel architecture, they are generally more efficient than CPUs. In some cases, the hardware platform 810 may include IaaS (Infrastructure as a Service) resources, which are computing resources (e.g., servers, memory, etc.) provided by a cloud service provider. The hardware platform 810 may also include computer memory for storing data related to the AI ​​model 830, applications for the AI ​​model 830, and training data for the AI ​​model 830. The computer memory can be in any form of random access memory (RAM), such as dynamic RAM, static RAM, or non-volatile RAM.

[0058] A software library 812 can be thought of as a suite of data and programming code, including executable files, used to control the computing resources of the hardware platform 810. The programming code may include low-level primitives (e.g., basic language elements) that form the basis of one or more low-level programming languages, allowing the servers of the hardware platform 810 to perform specific operations using these low-level primitives. Low-level programming languages ​​require little abstraction from the instruction set architecture of the computing resources, thus enabling them to run quickly with a small memory footprint. Examples of software libraries 812 that can be included in the AI ​​system 800 include the Intel MathKernel Library, Nvidia cuDNN, Eigen, and OpenBLAS.

[0059] The structural layer 804 may include a machine learning (ML) framework 814 and algorithms 816. The ML framework 814 can be thought of as an interface, library, or tool that enables users to build and deploy AI models 830. The ML framework 814 may include open-source libraries, application programming interfaces (APIs), gradient boosting libraries, ensemble methods, and / or deep learning toolkits that work in conjunction with layers of the AI ​​system to facilitate the development of AI models 830. For example, the ML framework 814 can distribute the process of applying or training AI models 830 across multiple resources within the hardware platform 810. The ML framework 814 may also include a set of pre-built components that provide the functionality to implement and train AI models 830, enabling users to build and train AI models 830 using pre-built functions and classes. Thus, the ML framework 814 facilitates the data engineering, development, hyperparameter tuning, testing, and training of AI models 830.

[0060] Examples of ML frameworks or libraries that can be used with AI System 800 include TensorFlow, PyTorch, Scikit-Learn, Keras, and Cafffe. Random Forest is a machine learning algorithm that can be used within ML Framework 814. LightGBM is a gradient boosting framework / algorithm (ML technique) that can be used. Other available techniques / algorithms include XGBoost and CatBoost. Amazon Web Services is a cloud service provider that offers a variety of machine learning services and tools (e.g., SageMaker) that can be used to build platforms, train, and deploy ML models.

[0061] In some implementations, the ML framework 814 learns data representations by directly performing deep learning (also known as deep structured learning or hierarchical learning) on ​​the input data, rather than using task-specific algorithms. In deep learning, no explicit feature extraction is performed; features in the feature vectors are implicitly extracted by the AI ​​system 800. For example, the ML framework 814 can use a cascade of multiple layers of nonlinear processing units for implicit feature extraction and transformation. Each subsequent layer uses the output from the previous layer as input. Thus, the AI ​​model 830 can be trained in supervised mode (e.g., classification) and / or unsupervised mode (e.g., pattern analysis). The AI ​​model 830 can learn multiple levels of representations corresponding to different levels of abstraction, where these different levels form a hierarchy of concepts. In this way, the AI ​​model 830 can be configured to distinguish features of interest from background features.

[0062] Algorithm 816 is a systematic set of computer-executable operations used to generate output data from a set of input data, and can be written using pseudocode. Algorithm 816 can contain complex code that allows computing resources to learn from new input data and create new / modified outputs based on what they have learned. In some implementations, Algorithm 816 can be trained while running on computing resources on a hardware platform 810 to build an AI model 830. This training allows Algorithm 816 to make predictions or decisions without explicit programming. Once training is complete, Algorithm 816 runs on computing resources as part of the AI ​​model 830 to make predictions or decisions, improve the performance of computing resources, or perform tasks. Algorithm 816 can be trained using supervised learning, unsupervised learning, semi-supervised learning, and / or reinforcement learning.

[0063] By using supervised learning, algorithm 816 can be trained to learn patterns (e.g., mapping input data to output data) based on labeled training data. The training data can be labeled by an external user or operator. For example, a user can collect a set of training data by capturing application and / or service usage patterns, metadata, past communication sessions, etc. (as detailed in Figures 5 and 6). The user can train the AI ​​model 830 by labeling the training data based on one or more classes and then inputting the training data into algorithm 816. The algorithm determines how to label new data based on the labeled training data. The user can easily collect, label, and / or input data via the ML framework 814. In some cases, the user can convert the training data into a set of feature vectors for input to algorithm 816. Once training is complete, the user can test algorithm 816 with new data to determine whether algorithm 816 predicts accurate labels for the new data. For example, a user can test the accuracy of algorithm 816 using cross-validation, and if the cross-validation result falls below the accuracy threshold, they can retrain algorithm 816 with new training data.

[0064] Supervised learning can include classification and / or regression. Classification techniques are used when the input data to algorithm 816 is discrete, and they train algorithm 816 to identify categories of new observations based on training data. In other words, when learning with classification techniques, algorithm 816 takes training data labeled with categories (e.g., classes) and determines how the features observed in the training data (e.g., data features in Figures 5 and 6 such as data transmission frequency, data session duration, data exchange volume, and data burst timing) relate to categories (e.g., services and applications). Once training is complete, algorithm 816 can classify new data by analyzing the new data against features mapped to categories. Examples of classification techniques include boosting, decision tree learning, genetic programming, learning vector quantization, k-nearest neighbors (k-NN) algorithms, and statistical classification.

[0065] Regression techniques estimate the relationship between independent and dependent variables and are used when the input data to algorithm 816 is continuous. Regression techniques can be used to train algorithm 816 to predict or forecast relationships between variables. To train algorithm 816 with regression techniques, the user can select a regression method to estimate the model parameters. The user collects and labels the training data to be input to algorithm 816 so that algorithm 816 is trained to understand the relationship between data features and the dependent variable. Once training is complete, algorithm 816 can predict missing historical data or future outcomes based on the input data. Examples of regression methods include linear regression, multiple regression, logistic regression, regression tree analysis, least squares, and gradient descent. In one implementation example, regression techniques can be used, for example, to estimate and impute missing data for machine learning-based preprocessing operations.

[0066] In unsupervised learning, algorithm 816 learns patterns from unlabeled training data. In particular, algorithm 816 is trained to learn hidden patterns and insights in the input data, which can be used for data exploration or the generation of new data. Here, algorithm 816 does not have predefined outputs, unlike the labels that are output when training algorithm 816 using supervised learning. Another way to train algorithm 816 using unsupervised learning and discover the underlying structure of a dataset is to group the data based on similarity and represent the dataset in a compressed format. The wireless communication system 400 disclosed herein can identify patterns in received data using unsupervised learning.

[0067] Several techniques can be used in supervised learning: clustering, anomaly detection, and techniques for training latent variable models. Clustering techniques group data into different clusters containing similar data, such that other clusters contain data that is not similar. For example, during clustering, potentially similar data remains in groups with low or no similarity to other groups. Examples of clustering techniques include density-based methods, hierarchical-based methods, partitioning methods, and grid-based methods. For example, algorithm 816 can be trained as a k-means clustering algorithm, which divides n observations into k clusters such that each observation belongs to a cluster whose nearest mean serves as the cluster's prototype. Anomaly detection techniques are used to detect rare objects or events in the data that have never been seen before, without prior knowledge of those objects or events. Anomalies can include data that occurs rarely in a set, deviations from other observations, outliers that do not match the rest of the data, and patterns that do not follow clearly defined normal behavior. When using anomaly detection techniques, Algorithm 816 can be trained to be a separated forest, a local outlier (LOF) algorithm, or a k-nearest neighbor (k-NN) algorithm. Latent variable techniques relate observable variables to a set of latent variables. These techniques assume that observable variables are the result of an individual's position on the latent variables, and that after controlling for the latent variables, the observable variables have nothing in common. Examples of latent variable techniques available in Algorithm 816 include factor analysis, item response theory, latent profile analysis, and latent class analysis.

[0068] In some embodiments, the AI ​​system 800 trains the algorithm 816 of the AI ​​model 830 based on training data to correlate feature vectors with expected outputs in the training data. As part of training the AI ​​model 830, the AI ​​system 800 forms a training set of features and training labels by identifying a positive training set of features determined to have the problem characteristics, and in some embodiments, a negative training set of features that do not have the problem characteristics. The AI ​​system 800 trains the AI ​​model 830 by applying the ML framework 814. When applied to feature vectors, the ML framework 814 outputs an indication of whether the feature vectors have a particular Boolean characteristic or an estimate of a relevant desired characteristic(s), such as the probability that the feature vectors have a particular Boolean characteristic or an estimate of a scalar characteristic. The AI ​​system 800 can further apply dimensionality reduction (e.g., by linear discriminant analysis (LDA), PCA, etc.) to reduce the amount of data in the feature vectors to a smaller, more representative dataset.

[0069] The model layer 806 implements the AI ​​model 830 using data from the data layer and algorithms 816 and ML frameworks 814 from the structure layer 804, thereby enabling the decision-making capabilities of the AI ​​system 800. The model layer 806 includes the model structure 820, model parameters 822, loss function engine 824, optimizer 826, and regularization engine 828.

[0070] Model structure 820 describes the architecture of AI model 830 of AI system 800. Model structure 820 defines the complexity of the patterns / relationships represented by AI model 830. Examples of structures that can be used as model structure 820 include decision trees, support vector machines, regression analysis, Bayesian networks, Gaussian processes, genetic algorithms, and artificial neural networks (or simply neural networks). Model structure 820 can include multiple structural layers, multiple nodes (or neurons) in each structural layer, and activation functions for each node. The activation function for each node defines how the node transforms the received data into a data output. A structural layer can include an input layer for nodes that receive input data and an output layer for nodes that generate output data. Model structure 820 can include one or more hidden node layers between the input and output layers. Model structure 820 is an artificial neural network (or simply neural network) that connects the nodes in the structural layers to interconnect them. Examples of neural networks include feedforward neural networks, convolutional neural networks (CNNs), recurrent neural networks (RNNs), autoencoders, and generative adversarial networks (GANs).

[0071] Model parameters 822 represent relationships learned during training and can be used to make predictions and decisions based on input data. Model parameters 822 can apply weights and biases to the nodes and connections of the model structure 820. For example, if the model structure 820 is a neural network, model parameters 822 can apply weights and biases to the nodes of each layer of the neural network, such that the weights determine the strength of the nodes and the biases determine the threshold of the activation function for each node. Model parameters 822, in conjunction with the node activation functions, determine how the input data is transformed into the desired output. Model parameters 822 can be determined and / or modified during the training of algorithm 816.

[0072] The loss function engine 824 can determine a loss function, which is a metric used to evaluate the performance of the AI ​​model 830 during training. For example, the loss function engine 824 can measure the difference between the predicted output of the AI ​​model 830 and the actual output of the AI ​​model 830, and this is used to guide the optimization of the AI ​​model 830 during training to minimize the loss function. The loss function can be presented via the ML framework 814, and the user can decide whether to retrain or modify the algorithm 816 if the loss function exceeds a threshold. In some cases, if the loss function exceeds a threshold, the algorithm 816 will be automatically retrained. Examples of loss functions include the binary cross-entropy function, hinge loss function, regression loss function (e.g., mean squared error, quadratic loss, etc.), mean absolute error function, smoothed mean absolute error function, log-cosh loss function, and quantile loss function.

[0073] The optimizer 826 adjusts the model parameters 822 to minimize the loss function during the training of the algorithm 816. In other words, the optimizer 826 uses the loss function generated by the loss function engine 824 as a guide to determine which model parameters lead to the most accurate AI model 830. Examples of optimizers include gradient descent (GD), adaptive gradient algorithm (AdaGrad), adaptive moment estimation (Adam), root mean square propagation (RMSprop), radial basis functions (RBF), and limited memory BFGS (L-BFGS). The type of optimizer 826 used can be determined based on the type of model structure 820, the data size, and the computational resources available in the data layer 802.

[0074] The regularization engine 828 performs regularization calculations. Regularization is a technique to prevent overfitting and underfitting of the AI ​​model 830. Overfitting occurs when the algorithm 816 is excessively complex and over-adapted to the training data, which can degrade the performance of the AI ​​model 830. Underfitting occurs when the algorithm 816 fails to recognize even basic patterns from the training data, resulting in poor performance on the training or validation data. The regularization engine 828 can apply one or more regularization techniques to properly fit the algorithm 816 to the training data. This helps to constrain the resulting AI model 830 and improve its general applicability. Examples of regularization techniques include Lasso (L1) regularization, Ridge (L2) regularization, and Elastic (L1 and L2) regularization.

[0075] In some embodiments, the AI ​​system 800 may include a feature extraction module implemented using components of an exemplary computer system 900, which are illustrated and described in more detail with reference to Figure 9. In some embodiments, the feature extraction module extracts a feature vector from the input data. The feature vector contains n features (e.g., feature a, feature b, ..., feature n). The feature extraction module reduces redundancy (e.g., repeated data values) in the input data to transform the input data into a reduced set of features, such as a feature vector. Since the feature vector contains relevant information from the input data, the AI ​​model 830 can use this reduced representation to identify thresholds for events or data values ​​of interest. In some exemplary implementations, the feature extraction module uses dimensionality reduction techniques: independent component analysis, Isomap, kernel principal component analysis (PCA), latent semantic analysis, partial least squares, PCA, multifactor dimensionality reduction, nonlinear dimensionality reduction, multilinear PCA, multilinear subspace learning, positive semidefinite embedding, autoencoders, and deep feature synthesis. Computer system

[0076] Figure 9 is a block diagram showing an exemplary computer system 900 according to one or more embodiments. In some examples, the computer system 900 includes one or more central processing units ("processors") 902, main memory 906, non-volatile memory 910, a network adapter 912 (e.g., a network interface), a video display 918, input / output devices 920, a control unit 922 (e.g., a keyboard and pointing device), a drive unit 924 including a storage medium 926, and a signal generation device 930, all connected to a bus 916. The bus 916 is illustrated as an abstraction representing one or more physical buses and / or point-to-point connections connected by appropriate bridges, adapters, or controllers. Therefore, in some examples, bus 916 includes the system bus, PCI (Peripheral Component Interconnect) bus or PCI-Express bus, HyperTransport or ISA (Industry Standard Architectural Interface) bus, SCSI (Small Computer System Interface) bus, USB (Universal Serial Buses), I2C (I2C) bus, or IEEE (Institute of Electrical and Electronics Engineers) standard 1394 bus (also known as "Firewire").

[0077] In some examples, computer system 900 shares a computer processor architecture similar to that of a computer, such as a desktop computer, tablet computer, PDA (Personal Digital Assistant), for example, a mobile phone, game console, music player, wearable electronic device (e.g., a wristwatch or fitness tracker), network-connected ("smart") device (e.g., a television or home assistant device), virtual / augmented reality system (e.g., a head-mounted display), or other electronic device that can execute a series of (sequential or otherwise) instructions specifying the actions that computer system 900 should perform.

[0078] Although the main memory 906, non-volatile memory 910, and storage medium 926 (also called “machine-readable medium”) are shown as a single medium, the terms “machine-readable medium” and “storage medium” should be interpreted to include a single or multiple medium (e.g., a centralized / distributed database and / or associated cache and server) that stores one or more instruction sets 928. The terms “machine-readable medium” and “storage medium” include any medium that can store, encode, or transport instruction sets executed by the computer system 900.

[0079] Generally, routines performed to implement embodiments of the present disclosure may be implemented as part of an operating system, or as a specific application, component, program, object, module, or instruction sequence (collectively referred to as a “computer program”). A computer program typically includes one or more instructions (e.g., instructions 904, 908, 928) that are set at various points in time in different memory and storage devices within a computing device. When read and executed by one or more processors 902, the instructions cause the computer system 900 to perform operations to execute elements, including various aspects of the present disclosure.

[0080] Furthermore, although the embodiments are described in the context of a fully functional computing device, those skilled in the art will understand that various embodiments can be distributed as various forms of program products. This disclosure applies regardless of the specific type of machine or computer-readable medium used to actually make the distribution.

[0081] Further examples of machine-readable storage media, machine-readable media, or computer-readable media include volatile and non-volatile memory 910, floppy disks and other removable disks, hard disk drives, optical discs (e.g., compact disc read-only memory (CD-ROM), digital multipurpose disc (DVD)), and transmission media for digital and analog communication links.

[0082] The network adapter 912 enables the computer system 900 to mediate data within the network 914 with external entities of the computer system 900 via any communication protocol supported by the computer system 900 and external entities. In some examples, the network adapter 912 includes network adapter cards, wireless network interface cards, routers, access points, wireless routers, switches, multi-layer switches, protocol converters, gateways, bridges, bridge routers, hubs, digital media receivers, and / or repeaters.

[0083] In some embodiments, the network adapter 912 includes a firewall that controls and / or manages permission for access / proxy to data within the computer network and tracks various levels of trust between different machines and / or applications. In some examples, the firewall is any number of modules having any combination of hardware and / or software components that can enforce a predetermined set of access rights between specific machines and applications, machines and machines, and / or applications and applications (e.g., regulating the flow of traffic and resource sharing between these entities). In some embodiments, the firewall also manages and / or accesses access control lists that show details of permissions, including the permissions for individuals, machines, and / or applications to access and operate objects, and the circumstances under which those permissions apply.

[0084] According to several embodiments, the technologies described herein are implemented by programmable circuits (e.g., one or more microprocessors), software and / or firmware, dedicated hardwired (i.e., non-programmable) circuits, or a combination of these. In some examples, the dedicated circuits are in the form of one or more application-specific integrated circuits (ASICs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), etc.

[0085] The description and drawings in this specification are illustrative and should not be construed as limiting. Numerous specific details are provided to enable a full understanding of this disclosure. However, known details may be omitted to avoid complicating the description. Furthermore, various modifications are possible without departing from the scope of the embodiments.

[0086] The terms used herein generally have their ordinary meanings in the art, in the context of this disclosure, and in the specific context in which each term is used. Specific terms used to describe this disclosure are described above or elsewhere in this specification to provide practitioners with additional guidance regarding the description of this disclosure. For convenience, certain terms may be highlighted, for example, using italics and / or quotation marks. The use of highlighting does not affect the scope and meaning of the terms. The scope and meaning of the terms are the same in the same context, with or without highlighting. It will be understood that the same thing can be expressed in multiple ways. It will be recognized that “memory” is a form of “storage,” and that these terms may be used interchangeably in some cases.

[0087] Accordingly, alternative languages ​​and synonyms may be used for any one or more of the terms described herein, but no special significance is placed on whether or not a term is described in detail herein. Synonyms for specific terms are provided. The enumeration of one or more synonyms does not preclude the use of other synonyms. Any examples in this specification, such as examples of terms discussed herein, are for illustrative purposes only and do not further limit the scope and meaning of this disclosure or any examples of terms illustrated herein. Similarly, this disclosure is not limited to the various embodiments described herein.

[0088] The embodiments and modifications shown and described herein are merely illustrative of the principles of the present invention, and it should be understood that various modifications can be made by those skilled in the art.

Claims

1. A method for locally cleaning the print head of a printing system, which includes: A step of performing a print head health check operation on the print head of the printing system, The print head health check operation includes identifying and recording a set of nozzle defects in a set of printer nozzles located within the print head. The nozzle defects of the set of nozzle defects include any of the following: clogging, misfiring, and mismatch of the set of printer nozzles in the print head. Step; A step of determining a set of nozzle degradation metric values ​​for the print head of the printing system, which indicates a set of nozzle defects in a set of printer nozzles, based on the health check operation of the print head, The set of nozzle degradation metric values ​​is associated with the health status of the printhead nozzle. Step; A step of defining a set of actions for the carrier associated with the print head of the printing system based on the set of nozzle degradation metric values, The carrier is a movable carriage fixed to the base of the printing system and equipped with a cleaning device configured to clean the print head, and The set of actions adjusts at least one parameter of the carrier or the cleaning device; Step; and A step of cleaning the print head according to the set of actions, A method for providing this.

2. The health check operation for the print head is as follows: A step of evaluating the printed template using one or more camera sensors of the printing system; Based on the evaluation, the step of identifying the cleanliness level of the print head using the carrier camera pointed at the plate surface of one or more printer nozzles in the set of printer nozzles; Using the carrier laser emitter-receiver pair, the steps include detecting a set of nozzle misses in response to the beam of the laser emitter-receiver pair crossing the expected trajectory of a droplet ejected from one or more printer nozzles in the set of printer nozzles of the print head: and A step of evaluating abnormal injection using the nozzle actuator of the carrier, The method according to claim 1, comprising any one of the following.

3. The method according to claim 1, wherein the identification and recording of the nozzle defects comprises identifying and recording the frequency, severity, and location of the nozzle defects within the printer nozzle.

4. The set of nozzle degradation metric values ​​is: The number of defective nozzles per unit area; The average severity of defects across the entire print head; and The distribution pattern of the nozzle defects in the print head, The method according to claim 1, wherein the method is one of the following.

5. The printhead health check operation comprises the step of capturing and analyzing a printed test pattern using an optical sensor or imaging device associated with the printing system, and The light sensor or imaging device is configured to detect deviations or defects in the print output printed using the print head. The method according to claim 1.

6. The aforementioned set of actions is: A step of adjusting the carrier speed based on the density of nozzle defects within a specific location of the print head; The steps of modifying the carrier's movement profile based on the density of the nozzle defects within the specific location of the print head; and A step of correcting the vacuum level of a vacuum pad mounted on the carrier or the contact pressure of a wiper mounted on the carrier, based on the density of the nozzle defects within the specific location of the print head. The method according to claim 1, comprising any of the following.

7. The adjusted at least one parameter of the carrier is: The speed of the carrier moving across the print head, The velocity of the carrier increases in response to a high density of nozzle defects within a particular location. speed; The carrier movement profile based on the distribution pattern of the nozzle defects within the print head; The vacuum level of the cleaning device, The vacuum level of the carrier increases in accordance with the high density of the nozzle defects in the particular location. Vacuum level; and The contact pressure of the wiper mounted on the carrier, The contact pressure of the wiper increases in accordance with the high density of the nozzle defects in the particular location. Contact pressure, The method according to claim 1, comprising any of the following.

8. When executed by at least one data processor in a computer system, the computer system: An operation to perform a health check operation on the print head of the printing system, The print head health check operation includes identifying and recording a set of nozzle defects in a set of printer nozzles located within the print head. The set of nozzle defects includes any of the following: clogging, misfiring, and mismatch of the set of printer nozzles in the print head. Action; An operation to determine a set of nozzle degradation metric values ​​for the print head of the printing system, which indicates a set of nozzle defects in the set of printer nozzles, based on the health check operation of the print head, The set of nozzle degradation metric values ​​is associated with the health status of the printhead nozzle. Action; An operation that generates a set of actions for the carrier associated with the print head of the printing system based on the set of nozzle degradation metric values, The carrier is a movable carriage fixed to the base of the printing system and equipped with a cleaning device configured to clean the print head, and The set of actions adjusts at least one parameter of the carrier or washing device. Operation; and An action to perform cleaning of the print head according to a set of actions, A non-temporary, computer-readable storage medium on which instructions for execution are recorded.

9. The non-temporary computer-readable storage medium according to claim 8, wherein the movement of the carrier is controlled by an actuator and an actuator controller.

10. By moving the carrier having the adjusted at least one parameter across the print head, the computer system can: Using a wiper, remove material from the plate surface of one or more printer nozzles in the set of printer nozzles. Using a vacuum pad, particles are removed from the plate surface of one or more printer nozzles, and A flushing fluid is sprayed onto the plate surface to remove one or more nozzle defects of the nozzle. The non-temporary computer-readable storage medium according to claim 8.

11. The health check operation for the print head is as follows: A step of evaluating the printed template using one or more camera sensors of the printing system; Based on the evaluation, the step of identifying the cleanliness level of the print head using the carrier camera pointed at the plate surface of one or more printer nozzles in the set of printer nozzles; Using the carrier laser emitter-receiver pair, the steps include detecting a set of nozzle outs in response to the beam of the laser emitter-receiver pair crossing the expected trajectory of a droplet ejected from one or more printer nozzles of the set of printer nozzles of the printhead; and A step of evaluating abnormal injection using the nozzle actuator of the carrier, A non-temporary computer-readable storage medium according to claim 8, comprising any of the above.

12. The non-temporary computer-readable storage medium according to claim 8, wherein the identification and recording of the nozzle defects includes any of the identification and recording of the frequency, severity, and location of the nozzle defects in the printer nozzle.

13. The set of nozzle degradation metric values ​​is Number of defective nozzles per unit area, The average severity of defects across the entire print head, and The distribution pattern of the nozzle defects in the print head, A non-temporary computer-readable storage medium according to claim 8, comprising any of the above.

14. The printhead health check operation comprises capturing and analyzing a printed test pattern using an optical sensor or imaging device associated with the printing system, and The optical sensor or imaging device is configured to detect any deviation or defect in the print output printed using the print head. The non-temporary computer-readable storage medium according to claim 8.

15. A print head comprising a set of printer nozzles, Each printer nozzle in the aforementioned set of printer nozzles is configured to print an image. Printhead: A carrier fixed to the base of the printing system, The carrier is a movable carriage, The carrier comprises a cleaning device configured to clean the print head. carrier; and A processing circuit configured to perform the following operations, An operation to perform a health check on the print head, The printhead health check operation includes identifying and recording a set of nozzle defects in the set of arranged printer nozzles, The nozzle defects of the set of nozzle defects include any of the following: clogging, misfiring, and mismatch of the set of printer nozzles. Action; An operation to determine a set of nozzle degradation metric values ​​for the print head that indicate a set of nozzle defects in a set of printer nozzles, based on the health check operation of the print head, The set of nozzle degradation metric values ​​is associated with the health status of the printhead nozzle. Action; An operation to generate a set of actions for the carrier based on the set of nozzle degradation metric values, The set of actions adjusts at least one parameter of the carrier or washing device. Operation; and An action to clean the print head according to the set of actions, A processing circuit configured to perform the following: A printing system equipped with the following features.

16. The health check operation of the print head performed by the processing circuit is: A step of evaluating the printed template using one or more camera sensors of the printing system; Based on the evaluation, the step of identifying the cleanliness level of the print head using the carrier camera pointed at the plate surface of one or more printer nozzles in the set of printer nozzles; Using the carrier laser emitter-receiver pair, a step of detecting a set of nozzle misses in response to a beam of the laser emitter-receiver pair crossing the expected trajectory of a droplet ejected from one or more printer nozzles in the set of printer nozzles of the print head; A step of evaluating abnormal injection using the nozzle actuator of the carrier, The printing system according to claim 15, comprising:

17. The printing system according to claim 15, wherein the identification and recording of nozzle defects comprises identifying and recording the frequency, severity, and location of the nozzle defects within the printer nozzle.

18. A set of nozzle degradation metric values, Number of defective nozzles per unit area, The average severity of defects across the entire print head, and The distribution pattern of the nozzle defects in the print head, The printing system according to claim 15, wherein the system represents any of the following.

19. The printhead health check operation comprises the step of capturing and analyzing a printed test pattern using an optical sensor or imaging device associated with the printing system, and The printing system according to claim 15, wherein the optical sensor or imaging device is configured to detect any deviation or defect in the print output printed using the print head. The printing system according to claim 15.

20. The aforementioned set of actions A step of adjusting the carrier speed based on the density of nozzle defects within a specific location of the print head; The steps of modifying the carrier's movement profile based on the density of nozzle defects within the specific location of the print head; and A step of changing the vacuum level of a vacuum pad equipped on the carrier or the contact pressure of a wiper equipped on the carrier based on the density of nozzle defects in a specific location of the print head. The printing system according to claim 15, comprising any of the above.

Images