A real-time printing path adaptive adjustment method and system for digital printing

Abstract

This invention relates to the field of industrial digital printing technology, specifically disclosing a real-time adaptive adjustment method and system for the printing path in digital printing. The method involves acquiring the real-time deviation of the printing medium and splitting it into low-frequency and high-frequency deviation components in the frequency domain. The low-frequency component drives the printhead module through a motion channel to achieve large-stroke physical displacement compensation, while the high-frequency component is converted into a nozzle data shift control quantity through a data channel to achieve high-frequency instantaneous compensation. Simultaneously, the shift control quantity is filtered by moving average to obtain a DC bias estimate. When the bias estimate exceeds a threshold, atomic operations are performed in the insensitive motion state of the motion platform, synchronously adding the bias quantity to the motion platform position reference and deducting the accumulated value of the shift control quantity, maintaining the mathematical identity of the synthesized compensation quantity. This invention can seamlessly eliminate DC bias in the data channel, effectively release the dynamic compensation range, avoid misalignment and tearing of printed patterns, adapt to high-speed continuous digital printing operations, and improve printing stability and yield.

Description

Technical Field

[0001] This invention relates to the field of industrial digital printing technology, and in particular to a method and system for real-time adaptive adjustment of the printing path in digital printing. Background Technology

[0002] In industrial printing scenarios such as high-speed digital printing, flexible printing media will generate two types of deformation when transported at high speed: one is a slow overall drift with a large offset amplitude and stable change, and the other is a local wrinkle and instantaneous jitter with a small offset amplitude and rapid change.

[0003] The industry commonly employs a collaborative compensation architecture between the motion channel and the data channel to address this issue. The motion channel carries the printhead module to perform large-range low-frequency following, completing large stroke deviation compensation, while the data channel performs small-range high-frequency compensation through dynamic displacement of inkjet data, achieving rapid deformation response. However, this architecture has some technical problems during long-term continuous operation. Due to the non-ideal asymmetry of media deformation and the existence of slight system drift, the compensation amount of the data channel gradually accumulates a non-zero DC bias. This bias continuously encroaches on the limited mapping range of the data channel, which is constrained by the physical arrangement of the nozzles and the line buffer capacity. This causes the effective dynamic range of the data channel for handling instantaneous high-frequency fluctuations to continuously shrink. When combined with sudden fluctuations, it is very easy to reach the compensation boundary, ultimately causing misalignment and tearing of the printed pattern. Furthermore, the existing conventional solution is to periodically reset the compensation amount of the data channel. This method directly causes instantaneous jumps in the synthetic compensation amount, forming misalignments and seams that are perceptible to the human eye on the continuously printed pattern, failing to simultaneously achieve the release of the dynamic range of the data channel and the guarantee of print pattern continuity.

[0004] Therefore, there is an urgent need for a real-time adaptive adjustment method and system for digital printing paths to solve the above problems. Summary of the Invention

[0005] The purpose of this invention is to provide a real-time adaptive adjustment method for the print path in digital printing, comprising the following steps: The deviation quantity from the vision sensor is obtained, which represents the real-time offset of the printing medium at the detection point relative to the theoretical position. The deviation quantity is input into the frequency domain splitter and separated into low-frequency deviation component and high-frequency deviation component. The low-frequency deviation component is transmitted along the first compensation branch and converted into a position adjustment command for the motion platform, driving the motion platform carrying the nozzle module to perform physical displacement. The high-frequency deviation component is transmitted along the second compensation branch and converted into a nozzle data shift control quantity acting on the inkjet data sequence of the current print line to adjust the landing position of the ink droplets on the media surface. The shift control quantity is biased and its DC bias estimate is obtained. When the absolute value of the DC bias estimate exceeds the preset bias threshold, the motion state of the motion platform is determined, and the bias elimination sequence is executed until the motion platform is detected to be in a preset insensitive motion state. The bias elimination sequence is an atomic operation that acts on both the first compensation branch and the second compensation branch: the DC bias estimate is superimposed on the target position reference of the motion platform as an offset, and the accumulated value of the shift control amount is deducted accordingly, so as to release the dynamic compensation range of the second compensation branch while keeping the synthetic compensation amount constant.

[0006] Furthermore, the present invention also discloses a real-time adaptive adjustment system for the print path in digital printing, comprising: The acquisition module is used to acquire the deviation from the vision sensor. The deviation represents the real-time offset of the printing medium at the detection point relative to the theoretical position. The deviation is input into the frequency domain splitter and separated into low-frequency deviation components and high-frequency deviation components. The conversion module is used to transmit the low-frequency deviation component along the first compensation branch and convert it into a position adjustment command for the motion platform, thereby driving the motion platform carrying the nozzle module to perform physical displacement. The adjustment module is used to transmit the high-frequency deviation component along the second compensation branch and convert it into a nozzle data shift control quantity that acts on the inkjet data sequence of the current print line, so as to adjust the landing position of the ink droplets on the medium surface, and to perform bias monitoring on the shift control quantity to obtain its DC bias estimate. The execution module is used to determine the motion state of the motion platform when the absolute value of the DC bias estimate exceeds the preset bias threshold, and to execute the bias elimination sequence when the motion platform is detected to be in a preset insensitive motion state. The processing module is used for the bias elimination sequence to be an atomic operation, and simultaneously applied to the first compensation branch and the second compensation branch: the DC bias estimate is superimposed on the target position reference of the motion platform as an offset, and the accumulated value of the shift control amount is deducted accordingly, so as to release the dynamic compensation range of the second compensation branch while keeping the synthetic compensation amount constant.

[0007] Furthermore, the execution module includes: The reading unit is used to enable the system to enter a bias elimination waiting state when the absolute value of the DC bias estimate exceeds the bias threshold, and to continuously read the real-time motion status fed back by the motion platform. The determination unit is used to determine whether the motion platform is in a non-sensitive motion state. The non-sensitive motion state refers to a state in which a small jump in the reference origin does not affect the continuity of the final continuous printing pattern. If the insensitive motion state is detected within the preset maximum waiting time, it is determined that the execution condition is met, and the bias elimination sequence is prepared to be executed. If the insensitive motion state is not detected within the maximum waiting time, a smooth rate change process is superimposed on the uniform motion command of the motion platform to artificially create the insensitive motion state condition. The execution unit is used to execute the bias elimination sequence in the first control cycle after the condition is determined to be met.

[0008] This application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described method for real-time adaptive adjustment of the print path in digital printing.

[0009] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method for real-time adaptive adjustment of the print path in digital printing.

[0010] The beneficial effects of this application are as follows: Firstly, this invention can achieve seamless elimination of DC bias. The bias elimination operation is performed only in the non-sensitive motion window of the motion platform. Furthermore, through the mathematical identity design of the synthetic compensation quantity, it ensures that the printed pattern has no instantaneous jumps, misalignments, or visual seams, thus completely solving the pattern defect problem caused by periodic zeroing.

[0011] Secondly, the present invention can maintain the dynamic compensation range of the data channel for a long time and seamlessly transfer the accumulated DC bias to the motion channel, so that the data channel is always in the center area of ​​the mapping interval, avoiding compensation saturation due to bias accumulation and greatly reducing the risk of pattern tearing.

[0012] Third, by optimizing the dual-channel collaborative compensation logic, this invention takes into account both the large-stroke low-frequency compensation capability of the motion channel and the high-response high-frequency compensation characteristics of the data channel, thereby improving the overall compensation accuracy and response speed of the system under existing hardware conditions. Attached Figure Description

[0013] Figure 1 This is a schematic diagram of a method flow proposed in an embodiment of this application.

[0014] Figure 2 This is a schematic diagram of the system structure proposed in an embodiment of the present invention.

[0015] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0016] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0017] like Figure 1 As shown, this application provides a real-time adaptive adjustment method for the print path in digital printing, including the following steps: S1, acquire the deviation from the vision sensor, the deviation representing the real-time offset of the printing medium at the detection point relative to the theoretical position, input the deviation into the frequency domain splitter, and separate it into low-frequency deviation component and high-frequency deviation component; S2, the low-frequency deviation component is transmitted along the first compensation branch and converted into a position adjustment command for the motion platform, driving the motion platform carrying the nozzle module to perform physical displacement; S3, the high-frequency deviation component is transmitted along the second compensation branch and converted into a nozzle data shift control quantity acting on the inkjet data sequence of the current print line to adjust the landing position of the ink droplets on the media surface; at the same time, the shift control quantity is biased and its DC bias estimate is obtained. S4. When the absolute value of the DC bias estimate exceeds the preset bias threshold, the motion state of the motion platform is determined until the motion platform is detected to be in a preset insensitive motion state, and then the bias elimination sequence is executed. S5, the bias elimination sequence is an atomic operation that acts on both the first compensation branch and the second compensation branch: the DC bias estimate is superimposed on the target position reference of the motion platform as a fixed offset, and an equal amount of offset is deducted from the accumulated value of the shift control quantity, so as to release the dynamic compensation range of the second compensation branch while maintaining the mathematical identity of the synthetic compensation quantity.

[0018] As described in steps S1-S5 above, in industrial printing scenarios using digital printing, flexible printing media undergoes two types of deformation during high-speed transport: large-amplitude, stable overall gradual drift, and small-amplitude, rapidly changing local wrinkles and instantaneous jitter. The industry commonly employs a dual-channel collaborative compensation architecture with a motion channel and a data channel. The motion channel handles large-stroke, low-frequency compensation, while the data channel handles small-range, high-frequency compensation. However, the compensation range of the data channel is limited by the physical capacity of the printhead line buffer. The asymmetry of media deformation and the slight system drift cause the shift control quantity to continuously accumulate DC bias. This bias continuously encroaches on the effective dynamic range of the data channel, and when combined with instantaneous high-frequency fluctuations, it easily reaches the compensation boundary, ultimately causing misalignment or tearing of the printed pattern. Directly clearing the data channel would cause a jump in the synthetic compensation quantity, disrupting pattern continuity. Therefore, a compensation adjustment method that can be executed seamlessly and effectively release the dynamic range of the data channel is needed.

[0019] Traditional dual-channel compensation schemes only implement basic deviation shunting and compensation functions, lacking a DC bias monitoring mechanism and logic for seamless bias elimination. Some schemes use periodic zeroing, which directly causes visual defects in the pattern and fails to transfer the bias to the motion channel. The dynamic range of the data channel is quickly squeezed again, resulting in extremely poor stability during long-term continuous printing. This invention integrates the core steps of deviation shunting, dual-channel compensation, bias monitoring, timing determination, and synchronous bias elimination to form a complete adaptive adjustment closed loop. This ensures print quality while maintaining the long-term, undiminished compensation capability of the data channel.

[0020] This invention constructs a closed-loop compensation system that coordinates the motion channel and the data channel through five core components: frequency domain current splitting of deviation, dual-channel collaborative compensation, real-time bias monitoring, non-sensitive motion state determination, and synchronous atomic operation for imperceptible debiasing. Without disrupting the continuity of the printed pattern, it continuously releases the dynamic compensation range of the data channel that is squeezed by DC bias, fundamentally solving the problems of compensation failure and pattern misalignment and tearing caused by the accumulation of data channel bias in high-speed digital printing.

[0021] This invention can be applied to digital printers, which continuously transmit data at a speed of 1 meter per second. The system first collects media offset data in real time through a vision sensor, and separates the data into a low-frequency deviation component adapted to the motion platform and a high-frequency deviation component adapted to the data channel at a cutoff frequency of 15Hz. The low-frequency deviation component is compensated for with large-stroke physical displacement through the XY micro-motion platform, and the high-frequency deviation component is converted into a shift control quantity of ±5 pixel lines to achieve zero-delay high-frequency compensation. The system extracts the DC bias estimate of the shift control quantity in real time through a moving average filter of 200 printing line cycles. When the value exceeds the offset threshold of ±3 pixel lines, the system waits for the natural zero-speed window of the motion platform. If it does not appear within 30 seconds, it actively creates a non-sensitive motion window. Finally, through a single-cycle atomic operation, the system synchronously adjusts the position reference of the motion platform and the cumulative value of the shift control quantity of the data channel, keeping the mathematical identity of the synthetic compensation quantity, seamlessly transferring the DC bias to the motion channel, fully releasing the dynamic compensation range of the data channel, and ensuring continuous and stable operation of the system.

[0022] In one embodiment, step S1 specifically includes: S101, the vision sensor is a high-speed line scan camera, which captures the reference marks or textures on the printing medium in real time before the printing medium enters the printing area. The image is processed to extract the position of the mark points with sub-pixel precision and generate a real-time deviation amount that varies over time and includes the paper feed direction component. S102, the real-time deviation is fed into a frequency domain splitter consisting of a low-pass filter and a high-pass filter, wherein the cutoff frequency of the splitter is set according to the measured response bandwidth of the motion platform to ensure that the rate of change of the low-frequency deviation component is within the physical limit that the motion platform can follow. S103, through the low-pass filter, extract the components in the real-time deviation whose rate of change is lower than the cutoff frequency to form a low-frequency deviation component, which corresponds to the overall slow drift and macroscopic tension fluctuation of the medium. S104, through the high-pass filter, extract the components in the real-time deviation quantity whose rate of change is higher than the cutoff frequency to form a high-frequency deviation component. This component corresponds to rapid deformations such as local micro-folds, instantaneous jitter, or seam jumps in the medium.

[0023] As described in steps S101-S104 above, by using a high-speed linear array camera to collect printing media offset data in real time, and using a frequency domain splitter that matches the measured response bandwidth of the motion platform to accurately separate the low-frequency deviation component and the high-frequency deviation component, the printing media deformation is processed in a layered manner according to the characteristics of the compensation channel, providing accurate layered deviation data support for the subsequent collaborative compensation of the motion channel and the data channel as well as the elimination of inductive DC bias.

[0024] In industrial printing scenarios using digital printing, the printing medium, when continuously transported at a speed of 1 meter per second, simultaneously experiences two types of deformation: one is a gradual, stable overall drift of the medium with a large offset amplitude, and the other is a localized, rapid, and instantaneous jitter of the medium with a small offset amplitude. The motion platform carrying the printhead module has a fixed physical response bandwidth limitation, and can only follow low-frequency deformations with a rate of change within this bandwidth. It cannot provide an effective physical response to high-frequency, rapid deformations. If the unlayered deviation data is directly input into the compensation system, it will be impossible to simultaneously meet the requirements for large stroke compensation and high response speed compensation. Therefore, it is necessary to separate the deviation quantity representing the real-time offset of the medium into two types of deviation components that are adapted to the motion channel and the data channel, respectively.

[0025] Traditional printing deviation processing techniques do not perform frequency-domain layered processing of the deviation; instead, they input all deviation data into a single compensation channel. This leads to problems such as the motion platform being unable to handle high-frequency jitter due to insufficient response bandwidth, causing misalignment of the printed pattern, or the data channel failing to compensate due to limited compensation travel. By limiting the type of vision sensor, the accuracy of deviation extraction, the configuration of the frequency domain splitter, and the basis for setting the cutoff frequency, precise frequency-domain separation of the deviation is achieved, allowing the separated components to be matched with the compensation characteristics of the motion channel and the data channel respectively.

[0026] Step S101 employs a high-speed line scan camera as a vision sensor. This sensor captures reference marks or textures on the printing medium in real time before it enters the printing area. An image processing algorithm extracts the mark positions with sub-pixel precision, generating a real-time deviation that varies over time and includes a paper feed direction component. In this embodiment, the elastic blended fabric operates at a conveying speed of 1 meter per second. The high-speed line scan camera can perfectly match this speed to achieve real-time data acquisition. Sub-pixel precision extraction accurately captures minute shifts in the medium, providing highly accurate raw data for subsequent deviation separation. This is the fundamental data guarantee for achieving layer compensation and subsequent offset elimination.

[0027] Step S102 sends the aforementioned real-time deviation to a frequency domain shunt composed of a low-pass filter and a high-pass filter. The cutoff frequency of the shunt is set according to the measured response bandwidth of the motion platform. In this embodiment, the motion platform is an XY micro-motion platform, and its measured response bandwidth along the Y-axis is 15Hz. Therefore, the cutoff frequency of the shunt is fixed at 15Hz. This setting ensures that the rate of change of the low-frequency deviation component is within the physical limits that the motion platform can follow. By fixing the hardware structure of the frequency domain shunt and the rules for setting the cutoff frequency, the failure of the deviation separation result due to an unreasonable cutoff frequency setting can be completely avoided.

[0028] Step S103 extracts components with a rate of change below the 15Hz cutoff frequency from the real-time deviation using a low-pass filter, forming a low-frequency deviation component. This component directly corresponds to the overall gradual drift and macroscopic tension fluctuations of the medium. This step independently separates the deviation component suitable for physical displacement compensation of the motion platform, providing dedicated processing data for the first compensation branch, enabling the motion platform to compensate for this large stroke deviation through physical displacement.

[0029] Step S104 uses a high-pass filter to extract components from the real-time deviation whose rate of change is higher than the 15Hz cutoff frequency, forming a high-frequency deviation component. This component directly corresponds to rapid deformations such as localized micro-wrinkles, instantaneous jitter, or seam jumps in the medium. This allows for the independent separation of deviation components that can only be quickly compensated through the data channel, providing accurate high-frequency compensation data for the second compensation branch. This high-frequency component is the core data source for subsequent DC bias accumulation. Accurately separating this component provides the necessary data prerequisite for subsequent bias monitoring and non-intrusive elimination, while also preventing high-frequency deformation data from entering the motion channel and causing response lag, thus ensuring the overall compensation system's response speed and compensation accuracy.

[0030] In one embodiment, step S2 specifically includes: S201, the motion platform is an XY micro-motion platform that carries the printhead module, and the low-frequency deviation component is input to a PID controller for the paper feeding direction; In S202, the PID controller calculates in real time the desired speed or desired position of the motion platform in the corresponding axial direction according to the magnitude and variation trend of the low-frequency deviation component, and converts it into a position adjustment instruction for the drive motor; In S203, the motion platform responds to the position adjustment instruction to drive the entire nozzle module to perform a smooth physical displacement, so that on the macro spatial scale, the nozzle module is synchronized with the overall drift of the print medium, and the large-stroke low-frequency deviation is compensated at the physical reference level.

[0031] As described in the above steps S201 to S203, by inputting the low-frequency deviation component into the PID controller, the X-Y micro-motion platform carrying the nozzle module is driven to perform a smooth physical displacement, so that the nozzle module follows the overall slowly varying drift of the print medium on the macro spatial scale, completes the physical compensation of the large-stroke low-frequency deviation, reserves a complete dynamic compensation range for the data channel, and provides reference support of the motion channel for subsequent non-inductive offset elimination.

[0032] In the industrial printing scenario of digital printing, the low-frequency deviation component obtained through frequency domain splitting corresponds to the overall slowly varying drift of the medium and macro tension fluctuations. The variation rate of this component is lower than 15Hz, which is within the physical response bandwidth range of the X-Y micro-motion platform and has a large offset amplitude. Limited by the capacity of the nozzle line buffer, the data channel can only handle small-range high-frequency deviations and cannot undertake the compensation of large-stroke low-frequency deviations. If this part of the deviation is processed by the data channel, it will quickly occupy the dynamic mapping interval of the data channel, resulting in the data channel being unable to respond to instantaneous high-frequency fluctuations, and eventually causing misalignment or tearing of the printed pattern. Therefore, the motion channel is required to complete the exclusive compensation of this part of the deviation through physical displacement.

[0033] Traditional motion compensation technology only directly converts the deviation amount into a position instruction of the motion platform, without performing dynamic calculation according to the magnitude and variation trend of the deviation. The motion platform is prone to response lag or displacement overshoot, cannot accurately match the slowly varying drift of the medium, and does not form a clear division of labor with the data channel, resulting in low compensation efficiency. By defining the type of motion platform, the type of controller and the control logic, the motion channel can accurately and smoothly follow the low-frequency deviation, form a stable division of labor with the data channel, and complete the large-stroke deviation compensation from the physical level.

[0034] In step S201, the motion platform is an X-Y micro-motion platform carrying the nozzle module. For the paper feeding direction, the low-frequency deviation component is input to a PID controller. This embodiment is applied to a single-pass digital textile printer, the Y-axis of the X-Y micro-motion platform corresponds to the medium paper feeding direction, and the low-frequency deviation component with a variation rate lower than 15Hz obtained through frequency domain splitting is directly input to the PID controller matched with the Y-axis, which clarifies the corresponding relationship between the motion platform axis and the controller, and provides a matching basis of hardware and data for subsequent precise control.

[0035] In step S202, the PID controller calculates the desired speed or position of the motion platform along the corresponding axis in real time based on the magnitude and trend of the low-frequency deviation component, and converts it into a position adjustment command for the drive motor. In this embodiment, the PID controller performs real-time calculations based on the input low-frequency deviation component and outputs a position adjustment command adapted to the Y-axis drive motor. The command accuracy matches the sub-pixel level deviation extraction standard, which can avoid displacement lag or overshoot of the motion platform and ensure that the physical displacement rhythm is perfectly matched with the gradual drift of the medium.

[0036] In step S203, the motion platform responds to the position adjustment command, driving the entire printhead module to perform a smooth physical displacement. This ensures that, on a macroscopic spatial scale, the overall drift of the printhead module and the printing medium remains synchronized, compensating for low-frequency deviations during large strokes at the physical reference level. In this embodiment, the Y-axis of the XY micro-motion platform performs a smooth displacement according to the position adjustment command, compensating for the large stroke offset caused by changes in overall media tension. This compensation is undertaken independently by the motion channel and does not occupy the mapping range of the data channel. This preserves the complete dynamic range for the data channel to handle high-frequency wrinkles and instantaneous jitter, and also provides stable support for the physical reference translation during subsequent offset elimination.

[0037] In one embodiment, step S3 specifically includes: S301, the high-frequency deviation component is amplified and rounded, and then converted into a signed pixel-level shift control quantity, the effective range of which is limited to the data mapping range provided by the line buffer built into the nozzle. S302, when each printing line cycle is triggered, the shift control quantity directly acts on the inkjet data sequence of the current printing line. By dynamically adjusting the data reading start pointer or trigger timing of the line buffer, the ink droplets actually ejected from the nozzle produce a positional shift on the media surface that is equivalent to the high-frequency deviation component, thereby achieving zero-delay instant compensation. S303, Simultaneously perform bias monitoring: Maintain a moving average filter for the shift control quantity, filter the historical sequence of the shift control quantity in real time according to a preset window length, and separate the slowly changing DC component from the continuous shift control quantity data. This DC component is the DC bias estimate that is gradually accumulated due to the asymmetry of dielectric deformation and the slight drift of the system. S304, compare the absolute value of the DC bias estimate with a preset bias threshold. The bias threshold is set according to the physical capacity of the row buffer mapping interval. Its value is less than the maximum physical boundary of the mapping interval to reserve a safety margin and prevent the shift control quantity from touching the mapping boundary after superimposed instantaneous high-frequency fluctuations.

[0038] As described in steps S301-S304 above, zero-delay high-frequency compensation of the data channel is achieved by converting high-frequency deviation components into pixel-level shift control quantities. At the same time, the DC bias estimate of the shift control quantity is extracted in real time through moving average filtering and compared with a preset bias threshold to provide a precise triggering condition for subsequent bias elimination and avoid the data channel dynamic compensation range being continuously squeezed by DC bias.

[0039] In industrial printing scenarios using digital printing, the high-frequency deviation component obtained through frequency domain splitting corresponds to rapid deformations such as localized micro-wrinkles and instantaneous jitter in the medium. This component's rate of change exceeds 15Hz, surpassing the physical response bandwidth of the XY micro-motion platform. It cannot be compensated for by physical displacement through the motion channel and can only be compensated instantly by the data channel through nozzle data shifting. The compensation capability of the data channel is limited by the data mapping range of the printhead's built-in line buffer. Furthermore, the asymmetry of medium deformation and minute system drift cause the shift control quantity to gradually accumulate DC bias. This bias continuously occupies the mapping range, reducing the effective dynamic range of the data channel to handle instantaneous high-frequency fluctuations, ultimately leading to misalignment or tearing of the printed pattern. Therefore, bias monitoring and threshold judgment need to be performed simultaneously during the data compensation process.

[0040] Traditional data channel compensation technology only achieves nozzle data shifting and lacks a DC bias monitoring mechanism. This allows DC bias to accumulate indefinitely until the data channel compensation saturates. Furthermore, it lacks a bias threshold as a trigger for elimination, making it impossible to prevent compensation failure in advance. By defining the rules for generating shift control quantities, the real-time compensation execution method, the DC bias extraction method, and the bias threshold setting standard, this technology achieves rapid high-frequency deviation compensation while simultaneously enabling real-time monitoring and over-limit judgment of DC bias, providing an accurate trigger signal for subsequent imperceptible bias elimination.

[0041] Step S301 amplifies and rounds the high-frequency deviation component, converting it into a signed pixel-level shift control quantity. Its effective range is limited to the data mapping range provided by the printhead's built-in line buffer. In this embodiment, the high-frequency deviation component is processed into a shift control quantity of ±5 pixel lines. This range perfectly matches the data mapping range of the printhead's line buffer, ensuring both the high-frequency compensation capability and defining physical boundaries for the shift control quantity to avoid exceeding the hardware's capacity.

[0042] In step S302, when each printing line cycle is triggered, the shift control quantity directly acts on the inkjet data sequence of the current printing line. By dynamically adjusting the data read start pointer or trigger timing of the line buffer, the ink droplets actually ejected from the nozzles are offset on the media surface by a positional shift equivalent to the high-frequency deviation component, thereby achieving zero-delay real-time compensation. In this embodiment, the printing line frequency is approximately 1kHz, and the shift control quantity takes effect synchronously in each line cycle, which can instantly offset the offset caused by local media wrinkles and instantaneous jitter. The zero-delay characteristic can avoid pattern defects caused by high-frequency deformation compensation lag.

[0043] Step S303 synchronously performs bias monitoring. The system maintains a moving average filter for the shift control quantity, and performs real-time filtering on the historical sequence of the shift control quantity according to a preset window length. This filters out the slowly changing DC component from the continuous shift control quantity data. This DC component is the DC bias estimate that gradually accumulates due to the asymmetry of dielectric deformation and the system's minute drift. The formula for calculating the DC bias estimate is: ; Among them, the This represents the DC bias estimate, used to characterize the slowly accumulating DC component of the shift control input. This indicates the length of the moving average filter window. Indicates the current print line cycle number. This represents the shift control amount for the i-th printing line cycle. In this embodiment, the window length of the moving average filter is set to 200 line cycles, which can smoothly eliminate instantaneous fluctuation interference and accurately extract the slowly accumulated DC bias component, providing real and reliable estimation data for bias elimination.

[0044] Step S304 compares the absolute value of the DC bias estimate with a preset bias threshold. The bias threshold is set based on the physical capacity of the row buffer mapping interval, and its value is less than the maximum physical boundary of the mapping interval to reserve a safety margin and prevent the shift control quantity from touching the mapping boundary after being superimposed with instantaneous high-frequency fluctuations. In this embodiment, the row buffer mapping interval is ±5 pixels, and the bias threshold is set to ±3 pixels. This threshold is 60% of the mapping interval, reserving sufficient dynamic range to cope with sudden high-frequency fluctuations. When the DC bias estimate exceeds this threshold, the subsequent bias elimination process can be triggered to prevent the dynamic range of the data channel from being completely squeezed.

[0045] In one embodiment, step S4 specifically includes: S401, when the absolute value of the DC bias estimate exceeds the bias threshold, the system is triggered to enter the bias elimination waiting state and continuously read the real-time motion status fed back by the motion platform servo driver. S402, determine whether the motion platform is in a non-sensitive motion state. The non-sensitive motion state refers to the state in which a small jump in the reference origin does not have a perceptible effect on the continuity of the final continuous printed pattern. A typical non-sensitive motion state is the zero-speed window in which the motion platform experiences zero motion speed during the speed direction switching process when performing the return motion of reciprocating scanning printing. S403, if the insensitive motion state is detected within the preset maximum waiting time, it is determined that the execution condition is met, and the bias elimination sequence is immediately prepared to be executed; S404, if the non-sensitive motion state is not detected within the maximum waiting time, a tiny triangular acceleration / deceleration waveform is actively superimposed on the uniform motion command of the motion platform. The peak acceleration of this waveform is strictly limited to less than one-tenth of the normal printing allowable acceleration, thereby artificially creating an extremely smooth rate change process, and determining that the non-sensitive motion state condition is met during this process. S405, within the first control cycle after determining that the condition is met in step S403 or step S404, the bias elimination sequence is executed.

[0046] As described in steps S401-S405 above, by monitoring the real-time motion status of the motion platform, bias elimination is preferentially performed in the natural non-sensitive motion state. When the natural non-sensitive motion state is not detected for a long time, a non-sensitive motion state is actively created to ensure that the bias elimination operation is performed in a timely manner without destroying the continuity of the printed pattern, thereby improving the engineering robustness and universality of the non-sensitive bias elimination mechanism.

[0047] In industrial printing scenarios using digital printing, bias elimination must be initiated once the estimated DC bias exceeds the bias threshold. Directly shifting the reference origin during the sensitive motion state of a uniform-speed printing platform will result in perceptible misalignments and seams in the printed pattern. In actual continuous printing, the platform may remain in a uniform-speed printing segment for extended periods, preventing the natural emergence of a non-sensitive motion state. Continuous waiting leads to a sustained accumulation of DC bias, continuously encroaching on the dynamic compensation range of the data channel, ultimately causing pattern tearing defects. Therefore, it is necessary to simultaneously equip the platform with both natural window detection and active window creation mechanisms.

[0048] Traditional offset elimination techniques do not strictly limit the timing of execution, often employing periodic random execution or direct zeroing. This easily triggers operations during sensitive printing periods, disrupting pattern continuity. Furthermore, they lack a mechanism for proactively creating non-sensitive windows, making it unable to eliminate offsets promptly under long-term, uniform printing conditions, thus limiting their applicability in engineering scenarios. By clearly defining the rules for determining non-sensitive motion states, setting a maximum waiting time, and standardizing the proactive creation of non-sensitive windows, the timing of offset elimination can be precisely controlled, balancing timeliness of elimination with the integrity of the printed pattern.

[0049] In step S401, when the absolute value of the DC bias estimate exceeds the bias threshold, the system is triggered, enters the bias elimination waiting state, and continuously reads the real-time motion status fed back by the motion platform servo driver. In this embodiment, when the DC bias estimate exceeds the bias threshold of ±3 pixel rows, the system immediately starts the waiting process, and collects the speed feedback data of the Y-axis of the XY micro-motion platform in real time, providing real-time basis for the determination of non-sensitive motion states.

[0050] In step S402, the system determines whether the motion platform is in a non-sensitive motion state. A non-sensitive motion state refers to a state where a small jump in the reference origin will not have a perceptible impact on the continuity of the continuously printed pattern. The zero-speed window during the return motion of the reciprocating scanning printout of the motion platform is a typical non-sensitive motion state. In this embodiment, after the XY micro-motion platform completes a single-segment printing task on the Y-axis, it performs a return motion. The zero-crossing point where the speed changes from positive to negative is the zero-speed window. During this period, performing reference translation will not produce visual defects.

[0051] In step S403, if a non-sensitive motion state is detected within the preset maximum waiting time, the system determines that the execution conditions are met and immediately prepares to execute the bias elimination sequence. In this embodiment, the system presets a maximum waiting time of 30 seconds. When a zero-velocity window on the Y-axis is detected within 30 seconds, the system directly confirms that the execution conditions have been met without activating a backup mechanism.

[0052] In step S404, if no non-sensitive motion state is detected within the maximum waiting time, the system actively superimposes a tiny triangular acceleration / deceleration waveform onto the uniform motion command of the motion platform. The peak acceleration of this waveform is limited to less than one-tenth of the normal printing allowable acceleration, artificially creating a smooth rate change process, and determining that the non-sensitive motion state condition is met during this process. In this embodiment, when the motion platform continuously prints at a uniform speed of 1 m / s for more than 30 seconds, the system superimposes a smooth acceleration / deceleration command, which neither affects the printing effect nor fails to create a compliant execution window.

[0053] In step S405, within the first control cycle after determining that the condition is met in step S403 or S404, the system executes the bias elimination sequence. In this embodiment, the system starts the operation within the first print line control cycle after confirming the execution condition, ensuring accurate execution timing, avoiding further accumulation of DC bias, and ensuring the timeliness of operation execution.

[0054] In one embodiment, the step S5, which involves releasing the dynamic compensation range of the second compensation branch while maintaining the mathematical identity of the synthetic compensation amount, specifically includes: S501, the atomic operation is completed sequentially within one control cycle: First, the DC bias estimate at the current moment is used as a fixed coordinate offset and superimposed on the current target position reference of the motion platform at once, so that the physical position reference of the motion platform is translated by an equal amount, and the updated physical position reference is equal to the sum of the original target position reference and the DC bias estimate. S502, Next, the accumulated value of the shift control quantity is subtracted from the DC bias estimate by an equal amount, so that the accumulated value of the shift control quantity is zero or significantly reduced, thereby pulling the compensation reference of the second compensation branch back to the center region of its data mapping interval. S503, before and after the atomic operation is performed, the combined compensation amount of the first compensation branch and the second compensation branch remains mathematically constant: before the atomic operation is performed, the combined compensation amount is the sum of the original target position reference and the cumulative value of the shift control amount; after the atomic operation is performed, the combined compensation amount is the sum of the updated physical position reference and the cumulative value of the shift control amount after being returned to zero. Since the updated physical position reference already includes the deducted bias amount, and the cumulative value of the shift control amount has been deducted by an equal amount, the two changes cancel each other out, and the value of the combined compensation amount does not change before and after the operation. S504, thus, the absolute landing position of the ink droplets on the media surface does not change at the moment of offset elimination, and the continuity of the printed pattern is maintained. However, inside the system, the DC bias originally borne by the second compensation branch has been seamlessly transferred to the physical position reference of the first compensation branch. The dynamic range occupied by the second compensation branch is completely released and restored to a state that can continue to respond to subsequent high-frequency fluctuations.

[0055] As described in steps S501-S504 above, the target position reference of the motion channel and the cumulative value of the shift control quantity of the data channel are synchronously adjusted through atomic operations within a single control cycle. Under the premise of maintaining the mathematical equality of the combined compensation quantity of the first compensation branch and the second compensation branch, the DC bias accumulated by the data channel is seamlessly transferred to the motion channel, completely releasing the dynamic compensation range that the data channel is squeezed out, while ensuring that the landing point of the printed pattern has no instantaneous jump and no visual defects.

[0056] In industrial printing scenarios using digital printing, the shift control of the data channel is limited by the mapping range of ±5 pixel lines of the printhead line buffer. When the DC bias accumulates to the bias threshold of ±3 pixel lines, the data channel has only a small dynamic range left to cope with local wrinkles and instantaneous jitter in the media. If sudden high-frequency fluctuations are superimposed, the total compensation will reach the boundary of the mapping range, directly causing misalignment or tearing of the printed pattern. If the shift control is directly cleared, the combined compensation of the two compensation paths will change abruptly, causing the absolute landing point of the ink droplets to shift instantaneously, forming a seam that is perceptible to the human eye on the continuous pattern. Therefore, it is necessary to keep the combined compensation constant during the bias elimination process, while simultaneously completing the bias transfer and range release.

[0057] Traditional offset elimination methods only directly zero out the data channel offset without simultaneously adjusting the motion channel position reference. This can cause jumps in the synthesized compensation amount and pattern defects. Some solutions only adjust the data channel reference, failing to transfer the offset to the motion channel. Consequently, the dynamic range of the data channel cannot be fully released, and the offset quickly accumulates again. By synchronously modifying the two compensation references through atomic operations and clarifying the implementation logic of the mathematical identity of the synthesized compensation amount, offset elimination is guaranteed to be seamless in principle, while completely releasing the dynamic compensation capability of the data channel.

[0058] In step S501, atomic operations are completed sequentially within one control cycle. The system first uses the current DC bias estimate as a fixed coordinate offset and superimposes it onto the current target position reference of the motion platform, causing the physical position reference of the motion platform to shift by an equal amount. The updated physical position reference is equal to the sum of the original target position reference and the DC bias estimate. In this embodiment, the DC bias estimate is +3 pixels. The system superimposes this value onto the target position reference of the Y-axis of the XY micro-motion platform, and the physical position reference of the motion platform is simultaneously shifted forward by 3 pixels, completing the preparation for the transfer of the bias to the motion channel.

[0059] In step S502, the system then subtracts the accumulated value of the shift control quantity from the DC bias estimate by an equal amount, causing the accumulated value of the shift control quantity to return to zero or significantly decrease, thereby pulling the compensation reference of the second compensation branch back to the center region of its data mapping interval. In this embodiment, the accumulated value of the shift control quantity is subtracted by +3 pixels, and the accumulated value returns to near zero. The mapping interval space originally occupied by the bias in the data channel is completely released, restoring it to a state where it can respond to high-frequency fluctuations at full scale.

[0060] In step S503, before and after the atomic operation, the combined compensation amount of the first compensation branch and the second compensation branch remains mathematically constant. Before the atomic operation, the combined compensation amount is the sum of the original target position reference and the cumulative value of the shift control amount. After the atomic operation, the combined compensation amount is the sum of the updated physical position reference and the cumulative value of the shift control amount after zeroing. Since the updated physical position reference already includes the deducted bias amount, and the cumulative value of the shift control amount has been deducted by an equal amount, the two changes cancel each other out, and the value of the combined compensation amount does not change before and after the operation. In this embodiment, the original target position reference is superimposed with +3 pixels, and the cumulative value of the shift control amount is deducted with +3 pixels, and the sum of the two remains unchanged, mathematically ensuring that the combined compensation amount does not have any sudden changes.

[0061] In step S504, the absolute landing position of the ink droplets on the media surface does not change at the moment of bias elimination, thus maintaining the continuity of the printed pattern. The DC bias originally borne by the second compensation branch is seamlessly transferred to the physical position reference of the first compensation branch. The dynamic range occupied by the second compensation branch is completely released, restoring it to a state where it can continue to respond to subsequent high-frequency fluctuations. In this embodiment, the peak utilization rate of the data channel offset is always below 65% after bias elimination, which can continuously cope with instantaneous media jitter. Even after several hours of continuous operation, there will be no pattern defects caused by data channel saturation, thus ensuring long-term stable printing quality.

[0062] like Figure 2 As shown, the present invention also discloses a real-time adaptive adjustment system for digital printing paths, comprising: The acquisition module is used to acquire the deviation from the vision sensor. The deviation represents the real-time offset of the printing medium at the detection point relative to the theoretical position. The deviation is input into the frequency domain splitter and separated into low-frequency deviation components and high-frequency deviation components. The conversion module is used to transmit the low-frequency deviation component along the first compensation branch and convert it into a position adjustment command for the motion platform, thereby driving the motion platform carrying the nozzle module to perform physical displacement. The adjustment module is used to transmit the high-frequency deviation component along the second compensation branch and convert it into a nozzle data shift control quantity that acts on the inkjet data sequence of the current print line, so as to adjust the landing position of the ink droplets on the medium surface, and to perform bias monitoring on the shift control quantity to obtain its DC bias estimate. The execution module is used to determine the motion state of the motion platform when the absolute value of the DC bias estimate exceeds the preset bias threshold, and to execute the bias elimination sequence when the motion platform is detected to be in a preset insensitive motion state. The processing module is used for the bias elimination sequence to be an atomic operation, and simultaneously applied to the first compensation branch and the second compensation branch: the DC bias estimate is superimposed on the target position reference of the motion platform as an offset, and the accumulated value of the shift control amount is deducted accordingly, so as to release the dynamic compensation range of the second compensation branch while keeping the synthetic compensation amount constant.

[0063] In one embodiment, the execution module includes: The reading unit is used to enable the system to enter a bias elimination waiting state when the absolute value of the DC bias estimate exceeds the bias threshold, and to continuously read the real-time motion status fed back by the motion platform. The determination unit is used to determine whether the motion platform is in a non-sensitive motion state. The non-sensitive motion state refers to a state in which a small jump in the reference origin does not affect the continuity of the final continuous printing pattern. If the insensitive motion state is detected within the preset maximum waiting time, it is determined that the execution condition is met, and the bias elimination sequence is prepared to be executed. If the insensitive motion state is not detected within the maximum waiting time, a smooth rate change process is superimposed on the uniform motion command of the motion platform to artificially create the insensitive motion state condition. The execution unit is used to execute the bias elimination sequence in the first control cycle after the condition is determined to be met.

[0064] This application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described method for real-time adaptive adjustment of the print path in digital printing.

[0065] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method for real-time adaptive adjustment of the print path in digital printing.

[0066] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in this application and in the embodiments can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0067] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, apparatus, article, or method. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.

[0068] The above description is merely a preferred embodiment of the present invention and does not limit the scope of this application. Any equivalent results or equivalent process transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the scope of protection of this application.

Claims

1. A method for real-time adaptive adjustment of the print path in digital printing, characterized in that, Includes the following steps: The deviation quantity from the vision sensor is obtained, which represents the real-time offset of the printing medium at the detection point relative to the theoretical position. The deviation quantity is input into the frequency domain splitter and separated into low-frequency deviation component and high-frequency deviation component. The low-frequency deviation component is transmitted along the first compensation branch and converted into a position adjustment command for the motion platform, driving the motion platform carrying the nozzle module to perform physical displacement. The high-frequency deviation component is transmitted along the second compensation branch and converted into a nozzle data shift control quantity acting on the inkjet data sequence of the current print line to adjust the landing position of the ink droplets on the media surface. The shift control quantity is biased and its DC bias estimate is obtained. When the absolute value of the DC bias estimate exceeds the preset bias threshold, the motion state of the motion platform is determined, and the bias elimination sequence is executed until the motion platform is detected to be in a preset insensitive motion state. The bias elimination sequence is an atomic operation that acts on both the first compensation branch and the second compensation branch: the DC bias estimate is superimposed on the target position reference of the motion platform as an offset, and the accumulated value of the shift control amount is deducted accordingly, so as to release the dynamic compensation range of the second compensation branch while keeping the synthetic compensation amount constant.

2. The real-time adaptive adjustment method for the printing path in digital printing according to claim 1, characterized in that, The step of inputting the deviation quantity into a frequency domain splitter to separate it into a low-frequency deviation component and a high-frequency deviation component includes: Before the printing medium enters the printing area, the vision sensor captures reference marks or textures on the medium in real time. The image processing extracts the position of the mark points and generates a real-time deviation that varies over time and includes a paper feed direction component. The real-time deviation is fed into a frequency domain splitter consisting of a low-pass filter and a high-pass filter. The cutoff frequency of the splitter is set according to the response bandwidth of the motion platform so that the rate of change of the low-frequency deviation component is within the range that the motion platform can follow. The low-frequency deviation component in the real-time deviation is extracted by low-pass filtering. This component corresponds to the overall gradual drift and macroscopic tension fluctuation of the medium. The high-frequency deviation component in the real-time deviation is extracted by high-pass filtering, and this component corresponds to the local rapid deformation of the medium.

3. The real-time adaptive adjustment method for the printing path in digital printing according to claim 1, characterized in that, The step of transmitting the low-frequency deviation component along the first compensation branch and converting it into a position adjustment command for the motion platform to drive the motion platform carrying the nozzle module to perform physical displacement includes: The motion platform is a micro-motion platform that carries the printhead module. For the paper feeding direction, the low-frequency deviation component is input to a controller. The controller calculates the motion control quantity of the motion platform in the corresponding axis in real time based on the magnitude and trend of the low-frequency deviation component, and converts it into a position adjustment command for the drive motor. The motion platform responds to the position adjustment command and drives the printhead module to perform physical displacement, keeping the overall drift of the printhead module and the printing medium synchronized on a macroscopic spatial scale, thus compensating for low-frequency deviations at the physical reference level.

4. The real-time adaptive adjustment method for the printing path in digital printing according to claim 1, characterized in that, The step of bias monitoring the shift control quantity and obtaining its DC bias estimate includes: The high-frequency deviation component is converted into a pixel-level shift control quantity, the effective range of which is limited by the data mapping range of the nozzle's built-in line buffer. When each print line cycle is triggered, the shift control quantity acts on the inkjet data sequence of the current print line. By dynamically adjusting the data reading start position or trigger timing of the line buffer, the ink droplets on the media surface generate a position offset corresponding to the high-frequency deviation component. Synchronous bias monitoring: The shift control quantity is filtered to separate the DC bias estimate that gradually accumulates due to the asymmetry of dielectric deformation and the slight drift of the system from the continuously obtained shift control quantity data; The absolute value of the DC bias estimate is compared with a preset bias threshold, which is set according to the physical capacity of the row buffer mapping interval and is less than the maximum physical boundary of the mapping interval.

5. The real-time adaptive adjustment method for the printing path in digital printing according to claim 1, characterized in that, The step of determining the motion state of the motion platform when the absolute value of the DC bias estimate exceeds a preset bias threshold, and executing the bias elimination sequence until the motion platform is detected to be in a preset insensitive motion state, includes: When the absolute value of the DC bias estimate exceeds the bias threshold, the system enters the bias elimination waiting state and continuously reads the real-time motion status fed back by the motion platform. Determine whether the motion platform is in a non-sensitive motion state. The non-sensitive motion state refers to a state in which a small jump in the reference origin does not affect the continuity of the final continuous printed pattern. If the insensitive motion state is detected within the preset maximum waiting time, it is determined that the execution condition is met, and the bias elimination sequence is prepared to be executed. If the insensitive motion state is not detected within the maximum waiting time, a smooth rate change process is superimposed on the uniform motion command of the motion platform to artificially create the insensitive motion state condition. Within the first control cycle after determining that the conditions are met, the bias elimination sequence is executed.

6. The real-time adaptive adjustment method for the printing path in digital printing according to claim 1, characterized in that, The step of adding the DC bias estimate as an offset to the target position reference of the motion platform and correspondingly deducting the accumulated value of the shift control amount, so as to release the dynamic compensation range of the second compensation branch while keeping the synthetic compensation amount constant, includes: Within a control cycle, the DC bias estimate at the current moment is superimposed on the current target position reference of the motion platform as an offset, causing the physical position reference of the motion platform to be translated. At the same time, the accumulated value of the shift control amount is deducted accordingly, so that it is restored to the center region of the mapping interval. Before and after the operation is performed, the combined compensation amount of the first compensation branch and the second compensation branch remains constant, and the absolute landing position of the ink droplets on the medium surface does not change at the moment of offset elimination. While maintaining the continuity of the printed pattern, the DC bias accumulated in the second compensation branch is transferred to the first compensation branch, thereby releasing the dynamic compensation range occupied by the second compensation branch.

7. A real-time adaptive adjustment system for the print path in digital printing, characterized in that, include: The acquisition module is used to acquire the deviation from the vision sensor. The deviation represents the real-time offset of the printing medium at the detection point relative to the theoretical position. The deviation is input into the frequency domain splitter and separated into low-frequency deviation components and high-frequency deviation components. The conversion module is used to transmit the low-frequency deviation component along the first compensation branch and convert it into a position adjustment command for the motion platform, thereby driving the motion platform carrying the nozzle module to perform physical displacement. The adjustment module is used to transmit the high-frequency deviation component along the second compensation branch and convert it into a nozzle data shift control quantity that acts on the inkjet data sequence of the current print line, so as to adjust the landing position of the ink droplets on the medium surface, and to perform bias monitoring on the shift control quantity to obtain its DC bias estimate. The execution module is used to determine the motion state of the motion platform when the absolute value of the DC bias estimate exceeds the preset bias threshold, and to execute the bias elimination sequence when the motion platform is detected to be in a preset insensitive motion state. The processing module is used for the bias elimination sequence to be an atomic operation, and simultaneously applied to the first compensation branch and the second compensation branch: the DC bias estimate is superimposed on the target position reference of the motion platform as an offset, and the accumulated value of the shift control amount is deducted accordingly, so as to release the dynamic compensation range of the second compensation branch while keeping the synthetic compensation amount constant.

8. The real-time adaptive printing path adjustment system for digital printing according to claim 7, characterized in that, The execution module includes: The reading unit is used to enable the system to enter a bias elimination waiting state when the absolute value of the DC bias estimate exceeds the bias threshold, and to continuously read the real-time motion status fed back by the motion platform. The determination unit is used to determine whether the motion platform is in a non-sensitive motion state. The non-sensitive motion state refers to a state in which a small jump in the reference origin does not affect the continuity of the final continuous printing pattern. If the insensitive motion state is detected within the preset maximum waiting time, it is determined that the execution condition is met, and the bias elimination sequence is prepared to be executed. If the insensitive motion state is not detected within the maximum waiting time, a smooth rate change process is superimposed on the uniform motion command of the motion platform to artificially create the insensitive motion state condition. The execution unit is used to execute the bias elimination sequence in the first control cycle after the condition is determined to be met.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images