Preparation method of PCB (printed circuit board) with high carbon ink

Abstract

The invention provides a preparation method of a high-carbon ink PCB. The preparation method comprises the following steps: acquiring a conductivity distribution data set containing a region identifier, a conductivity value and a boundary coordinate; a micron-order precision control algorithm is adopted to generate a motion control instruction set corresponding to the timestamp, the nozzle position and the deposition coordinate; acquiring a printing ink deposition image data set in the deposition process, and calculating a deviation value between the actual deposition amount and the target deposition amount; the deposition process is controlled according to the updated jet frequency configuration file, and the transition width of the boundary area is detected; if the transition width exceeds a micron-order threshold value, generating an ink deposition data set containing optimized boundary coordinates; based on the printing ink deposition data set, partition heating control parameters corresponding to the area identification and the curing temperature are generated; and measuring a final conductivity parameter and generating a verification data set corresponding to the region identifier and the actually measured conductivity. According to the invention, high-precision and controllable deposition of the high-carbon ink on the PCB is realized, and the conductivity and consistency of the high-carbon ink are effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of PCB board preparation, and in particular to a method for preparing a PCB board with high carbon ink. Background Art

[0002] Printed circuit board (PCB) manufacturing is a core area of the electronics industry, directly impacting the performance, reliability, and miniaturization of electronic products. With the rapid development of smart devices and flexible electronics, PCBs are required to implement complex functions within limited space. Regionalized design with differentiated conductive properties has become a key technical direction. This technology can enhance the integration and functional diversity of circuit boards and is of great significance to the development of 5G communications, the Internet of Things, and wearable devices. However, traditional PCB manufacturing methods have significant limitations in achieving regionalized conductive properties.

[0003] Currently, conventional PCB manufacturing relies primarily on etching and electroplating processes, which struggle to achieve differentiated conductive properties across different regions on the same board. Traditional processes typically utilize a single conductive material, making it difficult to precisely partition high and low conductive areas within complex circuit designs. Furthermore, existing printing technologies struggle with nozzle switching accuracy and ink deposition volume control during multi-material deposition, resulting in blurred conductivity boundaries between regions and impacting circuit performance stability. During the curing process, a single heating method also struggles to achieve optimal curing conditions for different conductive materials, further limiting manufacturing accuracy and reliability. The core challenge in fabricating PCBs containing high-carbon inks lies in achieving high-precision regional ink deposition and clear demarcation of differentiated conductive properties. Key technical factors include micron-level nozzle switching control within multi-nozzle printing equipment, the ability to monitor and adjust ink deposition volume in real time, and the adaptability of the zoned heating curing process to different conductive inks. Unresolved challenges such as these hinder precise control of regional conductivity, hindering the manufacture of high-performance PCBs.

[0004] Therefore, how to achieve micron-level precise deposition, clear boundaries and differentiated curing of high and low conductive areas through the regional printing process of high-carbon ink on the same PCB board has become a key issue in this study. Summary of the Invention

[0005] The present invention provides a method for preparing a PCB board with high carbon ink, which mainly comprises:

[0006] Obtaining a conductive property distribution data set including region identification, conductive property values, and boundary coordinates;

[0007] Generating a mapping relationship between nozzle number, ink type and target area coordinates according to the conductive performance distribution data set;

[0008] A micron-level precision control algorithm is used to generate motion control instruction sets corresponding to timestamps, nozzle positions, and deposition coordinates;

[0009] Acquire an ink deposition image data set during the deposition process and calculate a deviation between an actual deposition amount and a target deposition amount;

[0010] If the deviation value exceeds a preset threshold, updating an injection frequency profile including proportional-integral-derivative parameters;

[0011] controlling the deposition process based on the updated jet frequency profile and detecting the transition width in the boundary region;

[0012] If the transition width exceeds a micrometer threshold, generating an ink deposition dataset comprising optimized boundary coordinates;

[0013] generating a zone heating control parameter corresponding to a zone identifier and a curing temperature based on the ink deposition data set;

[0014] Acquire actual temperature distribution data during the curing process and generate new power control instructions if the temperature deviation exceeds a threshold;

[0015] Measure the final conductive performance parameters and generate a validation dataset with region identification corresponding to the measured conductivity.

[0016] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0017] The present invention discloses a method for the precise deposition and preparation of high-carbon ink on a printed circuit board (PCB). The method first obtains conductive performance distribution data, generates a mapping relationship between the nozzle and the target area, and uses a micron-level precision control algorithm to generate motion instructions. During the deposition process, deviations are calculated through real-time image analysis, and the injection frequency is dynamically adjusted. For boundary areas, the transition width is detected and the deposition data is optimized. During curing, heating is performed in zones according to regional characteristics, and the temperature distribution is monitored in real time and the power is adjusted. Finally, the conductive performance is measured and a verification data set is generated. The present invention achieves high-precision, controllable deposition of high-carbon ink on a printed circuit board (PCB), effectively improving the conductive performance and consistency of the printed circuit board, and providing a new technical solution for the manufacture of high-performance electronic circuits. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a flow chart of a PCB board with high carbon ink and a preparation method thereof according to the present invention;

[0019] Figure 2 Schematic diagram of a PCB board with high carbon ink and a preparation method thereof according to the present invention;

[0020] Figure 3 This is another schematic diagram of a PCB board with high carbon ink and a preparation method thereof according to the present invention. DETAILED DESCRIPTION

[0021] The following will describe the technical solutions in the embodiments of the present invention in detail with reference to the accompanying drawings. The described embodiments are only a part of the embodiments of the present invention.

[0022] like Figure 1-3 The method for preparing a PCB board with high carbon ink in this embodiment may specifically include:

[0023] Step S101 : obtaining a conductivity distribution data set including region identifiers, conductivity values, and boundary coordinates.

[0024] An initial data set is obtained from the sensor network, the initial data set including region identifiers, conductive performance values, and boundary coordinates. Noise detection is performed on the initial data set. If noise is detected, a median filtering algorithm is used to process the data set to obtain a denoised data set. Based on the region identifiers and boundary coordinates in the denoised data set, a K-means clustering algorithm is used to divide the regions, the number of clusters is determined, and a region division result is obtained. The conductive performance value of each region in the region division result is extracted from the denoised data set to generate a performance distribution feature. The variance of the performance distribution feature is calculated. If the variance is greater than a preset threshold, a linear interpolation algorithm is used to smooth the conductive performance value to obtain a smoothed performance distribution. Based on the smoothed performance distribution and the region division result, a conductive performance distribution data set including region identifiers, performance values, and boundary coordinates is generated. The boundary coordinates of the conductive performance distribution data set and the initial data set are compared using a boundary coordinate consistency verification algorithm to determine the final conductive performance distribution data set.

[0025] For example, the initial data set collected by the sensor network includes region identifiers, conductivity values, and boundary coordinates. For example, a soil monitoring area is divided into three sub-areas: A, B, and C. The initial data set records: the conductivity values of area A are 50, 52, and 48, and the boundary coordinates are (0, 0)-

[0026] (10,10); the conductivity values of region B are 60, 65, and 70, and the boundary coordinates are (10,0)-(20,10); the conductivity values of region C are 45, 50, and 55, and the boundary coordinates are (0,10)-(10,20). This data may contain noise due to sensor jitter or environmental interference.

[0027] In one possible implementation, noise detection is achieved by checking for abnormal fluctuations in conductivity values. If the difference between values in a certain area exceeds a preset threshold (e.g., 10), it is determined to be noise.

[0028] For example, the difference between 65 and 70 in region B is 5, which does not exceed the threshold and is considered noise-free. However, if a value suddenly changes to 90, it is considered noise. After the noise is detected, the median filter algorithm is used to replace the outlier with the middle value of the adjacent data.

[0029] For example, if the value 48 in region A changes to 90, we replace it with the median value 52 of 50, 52, and 90 to obtain the denoised data set: region A is 50, 52, and 52. This method can effectively smooth out outliers, preserve data trends, and improve the reliability of subsequent analysis.

[0030] Specifically, based on the region identification and boundary coordinates of the denoised dataset, the K-means clustering algorithm is used to divide the regions. Assuming the initial number of clusters K = 3, the algorithm assigns data points to the nearest cluster based on the geometric center of the boundary coordinates.

[0031] For example, the coordinate centers of regions A, B, and C are (5,5), (15,5), and (5,15), respectively. After clustering, the original divisions are maintained, resulting in the region division results. The advantage of K-means clustering is that it can automatically optimize region boundaries, adapt to complex terrain, and improve division accuracy.

[0032] In one embodiment, the conductivity property value of each region is extracted from the denoised data set to generate a property distribution feature.

[0033] For example, the performance distribution characteristic has an average value of 51.3 in region A, 65 in region B, and 50 in region C. This characteristic reflects spatial variations in soil conductivity and can be used to assess soil fertility or salinity distribution. The variance of the performance distribution characteristic is calculated. For example, if the variance is 50, exceeding the preset threshold of 30 indicates significant fluctuations in the performance values, potentially affecting analytical stability.

[0034] Preferably, a linear interpolation algorithm is used to smooth the performance values.

[0035] For example, the conductivity values of 60, 65, and 70 in region B fluctuate significantly. Interpolation generates intermediate values of 62.5 and 67.5, resulting in a smoothed sequence of 60, 62.5, 67.5, and 70. Smoothing reduces the impact of sudden changes and makes the performance distribution more continuous, facilitating subsequent modeling and visualization.

[0036] For example, based on the smoothed performance distribution and regional division results, a conductive performance distribution dataset is generated, which includes region identification, performance value and boundary coordinates, such as region A: performance value 51.3, coordinates (0,0)-

[0037] (10,10). This dataset integrates spatial and performance information and can be used for soil management decision making.

[0038] It can be understood that the boundary coordinates of the conductive property distribution dataset and the initial dataset are compared by the boundary coordinate consistency verification algorithm to ensure data consistency.

[0039] For example, the coordinates (0,0)-(10,10) of region A are verified to be consistent. If deviations are found, the coordinates are adjusted to match the initial dataset. The resulting conductivity distribution dataset is highly accurate and consistent, enabling support for precision agriculture or environmental monitoring.

[0040] It's important to note that each step in the above method focuses on improving data quality and analytical accuracy. Noise detection and filtering ensure data reliability, clustering optimizes regional divisions, smoothing enhances distribution continuity, and boundary verification ensures spatial consistency. These technical effects collectively support the precise analysis of soil conductivity and provide a reliable basis for scientific decision-making.

[0041] It can be understood that this embodiment uses soil and farmland as examples to describe the method adopted in this application and the specific application of high-carbon ink on PCB boards. Therefore, the above content related to soil and farmland is only for illustrative purposes and is only used to verify and explain the methods and means adopted in this application, and does not deviate from the subject content of this application.

[0042] Step S102: generating a mapping relationship between the nozzle number, ink type and target area coordinates according to the conductive performance distribution data set.

[0043] Region identifiers, conductivity values, and boundary coordinates are obtained from a conductivity distribution dataset. A data extraction algorithm is used to generate a data table containing region identifiers, conductivity values, and boundary coordinates, resulting in an initial data table. Regions are then partitioned using a K-means clustering algorithm based on the region identifiers and boundary coordinates in the initial data table, resulting in a region partitioning result. If the boundary coordinates in the region partitioning result are inconsistent with those in the initial data table, a coordinate consistency verification algorithm is used to adjust the region partitioning result, resulting in a consistent region partitioning result. Region identifiers and target region coordinates for each region are extracted from the consistent region partitioning, generating a region coordinate table containing the region identifiers and target region coordinates, resulting in a region coordinate table. Based on the region identifiers in the region coordinate table, a preset nozzle allocation rule is used to match nozzle numbers and ink type names, generating a nozzle allocation table containing nozzle numbers, ink type names, and target region coordinates, resulting in a nozzle allocation table. The nozzle numbers, ink type names, and target region coordinates are then extracted from the nozzle allocation table, generating a mapping relationship table containing the nozzle numbers, ink type names, and target region coordinates, resulting in a mapping relationship table. According to the target area coordinates in the mapping relationship table and the boundary coordinate points in the consistent area division, the coordinate consistency verification algorithm is used to verify the mapping relationship table to obtain the final mapping relationship table.

[0044] For example, a conductivity distribution dataset contains region identifiers, conductivity values, and boundary coordinates to describe the spatial distribution of soil conductivity. For example, for a patch of farmland, the dataset records three regions: Region X has conductivity values of 40, 42, and 45, with boundary coordinates from (0, 0) to (8, 8); Region Y has conductivity values of 50, 55, and 60, with boundary coordinates from (8, 0) to (16, 8); and Region Z has conductivity values of 35, 38, and 40, with boundary coordinates from (0, 8) to (8, 16). The data extraction algorithm generates an initial data table from this dataset, containing region identifiers (such as X, Y, and Z), a sequence of conductivity values, and corresponding boundary coordinates. This table has a clear structure and is easy to use for subsequent processing.

[0045] In one possible implementation, the K-means clustering algorithm is used to partition regions based on the region identifiers and boundary coordinates in the initial data table. Assuming K = 3, the algorithm uses the geometric center of the boundary coordinates, such as (4,4) for region X, (12,4) for region Y, and (4,12) for region Z, as the initial cluster center, assigning data points to the closest cluster. If the boundary coordinates of region X become (0,0)-(9,9) after clustering, which is inconsistent with the initial data table, a coordinate consistency verification algorithm is used to compare boundary point deviations and adjust them to (0,0)-(8,8) to achieve a consistent region partitioning, ensuring accurate spatial partitioning.

[0046] Specifically, the region identifiers and target region coordinates are extracted from the consistent region divisions, and a region coordinate table is generated.

[0047] For example, the identifier of region X is X, and the target region coordinates are (0,0)-(8,8). This table provides the basis for subsequent nozzle assignment. The nozzle assignment rule matches the nozzle number value and ink type name based on the region identifier.

[0048] For example, area X matches nozzle S1, and the ink type is nutrient solution A; area Y matches nozzle S2, and the ink type is nutrient solution B. Generate a nozzle allocation table, including nozzle S1, nutrient solution A, coordinates (0,0)-

[0049] (8,8) and other information to clarify the correspondence between the nozzle and the area.

[0050] Preferably, the nozzle number value, ink type name and target area coordinates are extracted from the nozzle allocation table to generate a mapping relationship table.

[0051] For example, the mapping table records sprinkler S1's correspondence with nutrient solution A and coordinates (0,0)-(8,8). To ensure accuracy, a coordinate consistency verification algorithm is used to compare the target area coordinates in the mapping table with the boundary coordinates of the consistent area. If the coordinate deviation of sprinkler S2 is found to be (8,0)-(17,8), it is adjusted to (8,0)-(16,8), resulting in the final mapping table. This table supports precise spraying equipment scheduling.

[0052] It can be understood that the above process is progressive through data extraction, clustering, coordinate verification and mapping relationship generation, ensuring the high consistency and practicality of the data table from the beginning to the end.

[0053] For example, the initial data table provides complete information, the regional coordinate table focuses on spatial division, and the nozzle allocation table and mapping relationship table directly serve the spraying operation, with rigorous logic and interconnected links.

[0054] Step S103 : Using a micron-level precision control algorithm to generate a motion control instruction set corresponding to the timestamp, nozzle position, and deposition coordinates.

[0055] Timestamps and nozzle trajectory data are obtained from time series data. Threshold segmentation is used to separate the position coordinates within the nozzle trajectory data to obtain a set of nozzle position coordinates. Target deposition coordinates are acquired using a visual sensor to generate an initial coordinate dataset. Based on the timestamps and nozzle position coordinates in the initial coordinate dataset, a Kalman filter is used to calculate the motion acceleration of adjacent coordinate points. If the motion acceleration exceeds a preset acceleration threshold, a discontinuity in the nozzle trajectory is determined. Cubic spline interpolation is used between the discontinuities to generate additional coordinate points, resulting in a smoothed trajectory where the rate of change of velocity is below a preset smoothness threshold. The interpolated nozzle position coordinates are extracted from the smoothed trajectory and mapped to the target deposition coordinates using time window matching. A mapping table is generated between timestamps, nozzle position coordinates, and target deposition coordinates. The target deposition coordinates corresponding to each timestamp in the mapping table are matched to a preset G-code instruction template to generate a preliminary instruction set. The actual motion path during execution of the preliminary instruction set is measured using a laser interferometer. If the deviation between the actual motion path and the target deposition coordinates exceeds a preset tolerance threshold, the parameters of the preliminary instruction set are adjusted to obtain the final motion control instruction set.

[0056] For example, time series data typically includes timestamps and nozzle motion trajectory data, which are used to describe the position changes of nozzles during precision spraying in farmland.

[0057] For example, a set of time series data records the position coordinates of the nozzle at timestamp t1 = 10:00:00 as (2, 3), and at timestamp t2 = 10:00:01 as (2.5, 3.2). Through the threshold segmentation method, the position change threshold can be set to 0.3 to separate the coordinate points of continuous movement and obtain the nozzle position coordinate set, such as {(2, 3),

[0058] (2.5, 3.2)}. This process ensures the elimination of outliers in the data, providing a reliable basis for subsequent analysis.

[0059] In a possible implementation, a visual sensor is used to obtain target deposition coordinates.

[0060] For example, a sensor scans a field and identifies the target deposition coordinates as (3, 4). This generates an initial coordinate dataset containing timestamp t1 and coordinates (3, 4). This dataset provides a spatial basis for matching sprinkler movement with the target area, reducing the possibility of blind spraying.

[0061] Specifically, the Kalman filter is used to calculate the motion acceleration of adjacent coordinate points of the nozzle.

[0062] For example, based on the coordinates (2,3) and (2.5,3.2), combined with a timestamp interval of 1 second, the filter estimates the acceleration. If the preset acceleration threshold is 0.5m / s 2 , and the calculated result is 0.7m / s 2 , then it is determined that there is a discontinuity. This method improves the robustness of trajectory analysis through dynamic prediction and correction.

[0063] Preferably, cubic spline interpolation is used between the discontinuity points to generate newly added coordinate points.

[0064] For example, inserting the point (2.3,3.1) between (2,3) and (2.5,3.2) generates a smooth motion trajectory whose velocity change rate is lower than the smoothness threshold of 0.2m / s. 2 The smooth trajectory reduces the vibration of the nozzle and improves the uniformity of spraying.

[0065] For example, we extract interpolated coordinates from a smooth motion trajectory, such as (2.3, 3.1), and establish a mapping relationship with the target deposition coordinates (3, 4) through time window matching. This generates a mapping table that records the correspondence between t1 (2.3, 3.1) and (3, 4). This table provides precise time-space associations for subsequent command generation.

[0066] In one embodiment, a preliminary instruction set is generated by matching G-code instruction templates according to a mapping relationship table.

[0067] For example, for (3,4), the command "G1X3Y4F100" is generated, indicating that the nozzle moves to (3,4) at a speed of 100mm / s. This process converts abstract coordinates into instructions that can be executed by the device.

[0068] It will be appreciated that the laser interferometer measures the actual path of motion.

[0069] For example, after executing a command, the actual path is (3.1, 4.2), with a deviation of 0.22 from the target (3, 4), exceeding the tolerance threshold of 0.1. At this point, the command parameters are adjusted, such as reducing the speed to 80 mm / s, to generate the final motion control command set. This verification and adjustment mechanism ensures high precision in spraying operations.

[0070] It's important to note that the aforementioned method provides systematic support for precision spraying through a progressive approach involving time series analysis, trajectory smoothing, coordinate mapping, and command optimization. Each step is logically rigorous and mutually supportive, centered around the core requirements of field spraying.

[0071] Step S104 , obtaining an ink deposition image data set during the deposition process, and calculating a deviation between the actual deposition amount and the target deposition amount.

[0072] An industrial camera captures images of the ink deposition process at a preset frame rate and captures images at preset time intervals to generate a raw image set. An image segmentation algorithm is used to process the raw image set, with a preset number of iterations. The outlines of the ink deposition areas are extracted to generate a segmented image set. A contour area calculation function is used to calculate the pixel area of each contour area in the segmented image set, multiplied by a preset conversion coefficient to generate an actual deposition dataset. A target deposition dataset is generated by reading target deposition values from a configuration file. If the absolute difference between the actual and target deposition values exceeds a preset threshold, a mean filter function is used to filter the difference sequence with a preset window size to generate a deposition deviation dataset. The deposition deviation dataset is then matched against a pre-established compensation parameter comparison table to generate an adjustment parameter set; the compensation parameter comparison table contains a mapping between deviation ranges and compensation coefficients. Based on the compensation coefficients in the adjustment parameter set, a contour detection function is used to redetect the boundaries of the deposition areas in the segmented image set to generate an optimized deposition dataset.

[0073] For example, an industrial camera captures images of the ink deposition process at a preset frame rate, typically 30 frames per second, to ensure dynamic deposition details are captured. The camera is aimed at the inkjet area and captures a frame every 0.5 seconds to generate a raw image set.

[0074] For example, 20 frames of images can be generated within 10 seconds to record the continuous changes in ink deposition.

[0075] It should be noted that the selection of frame rate and interval needs to balance image quality and processing efficiency. High frame rate is suitable for fast deposition scenarios, while low frame rate is suitable for stable processes.

[0076] In one possible implementation, an image segmentation algorithm processes the original image set, often using a threshold-based segmentation method. The number of iterations is set to 5, and the ink area and background are separated through multiple optimizations.

[0077] For example, setting a grayscale threshold of 150 to extract the outline of the ink area generates a segmented image set. If the background is complex, an edge detection algorithm can be used to enhance the clarity of the outline. This segmented image set provides a reliable foundation for subsequent area calculations.

[0078] Specifically, the contour area calculation function statistically calculates the pixel area of each contour in the segmented image set.

[0079] For example, if the outline pixel area of a frame image is 10,000 pixels, multiplying it by the preset conversion coefficient of 0.01 square millimeters / pixel will give an actual deposition amount of 100 square millimeters. Repeat this process to generate an actual deposition amount dataset containing deposition amounts of multiple frames.

[0080] Preferably, the conversion coefficient needs to be calibrated according to the camera resolution and actual size to ensure data accuracy.

[0081] For example, if the target deposition volume per frame is 120 square millimeters, a target deposition volume dataset is generated by reading the target deposition volume value from a configuration file. If the absolute difference between the actual deposition volume of 100 square millimeters and the target of 120 square millimeters is 20 square millimeters, exceeding the preset threshold of 10 square millimeters, further processing is required. Configuration files can store multiple target values to meet different inkjet requirements.

[0082] In one embodiment, the mean filter function processes the difference sequence, and the window size is set to 3 frames.

[0083] For example, a difference sequence of 20, 18, and 22 square millimeters can be filtered to produce a smoothed sedimentation deviation dataset, such as 19.33 square millimeters. Filtering reduces the effects of noise and improves the stability of the deviation data. The window size can be adjusted based on the sedimentation rate; rapidly changing scenes require a smaller window.

[0084] It can be understood that the deposition deviation data set matches the compensation parameter comparison table, and the comparison table records the corresponding relationship between the deviation range and the compensation coefficient.

[0085] For example, a deviation of 10-20 square millimeters corresponds to a compensation coefficient of 1.1. After matching, an adjustment parameter set is generated, including the coefficient 1.1.

[0086] It should be noted that the comparison table needs to be pre-built based on experimental data to ensure the targeted compensation.

[0087] In one embodiment, the contour detection function re-detects the deposition area boundary of the segmented image set according to the compensation coefficient 1.1 of the adjustment parameter set.

[0088] For example, the detection threshold is adjusted, the contour is re-extracted, and an optimized sedimentation dataset is generated. For example, the sedimentation is adjusted to 118 square millimeters, which is close to the target value.

[0089] Preferably, morphological operations can be combined to smooth the boundaries and improve detection accuracy. The optimized data set provides an accurate basis for subsequent inkjet control.

[0090] In step S105 , if the deviation value exceeds a preset threshold, the injection frequency configuration file including the proportional-integral-derivative parameters is updated.

[0091] Obtain a sequence of deposition deviation values. If the deposition deviation value sequence exceeds a preset threshold, extract the time series characteristics of the deposition deviation value sequence using the ARIMA model to obtain the deviation change trend. Based on the deviation change trend, use the Kalman filter to smooth the deposition deviation value sequence to obtain a smoothed deviation sequence. If the fluctuation amplitude of the smoothed deviation sequence exceeds the preset amplitude threshold, use the normal distribution to fit the smoothed deviation sequence and determine the deviation distribution parameters. Based on the deviation distribution parameters, use the gradient descent method to update the proportional parameters, integral parameters, and differential parameters in the configuration file to obtain an updated control parameter set. Use the PID controller to calculate the injection frequency value based on the updated control parameter set to obtain an adjusted frequency parameter set. If the difference between the adjusted frequency parameter set and the historical frequency parameter set exceeds the preset difference threshold, use the root locus method to perform stability analysis on the adjusted frequency parameter set to obtain a verified frequency parameter set. Based on the verified frequency parameter set, use the JSON format to update the injection frequency configuration file to obtain the final frequency configuration file.

[0092] For example, a sequence of deposition deviation values can be acquired through a combination of image processing and data analysis. In an inkjet deposition scenario, an industrial camera captures an image of the inkjet area, which is then segmented and calculated to determine the actual deposition amount. Assuming the target deposition amount is 100 square millimeters and the actual value is 95 square millimeters, the deviation is 5 square millimeters. Multiple frames of images are continuously captured to generate a sequence of deviations, such as 5, 6, 4, and 7 square millimeters.

[0093] It should be noted that the deviation sequence needs to be updated regularly to reflect the dynamic changes of the inkjet process.

[0094] In one possible implementation, if the deviation series exceeds a preset threshold, such as 3 square millimeters, the ARIMA model is used to analyze the time series characteristics. The ARIMA model extracts trend and cyclical characteristics by analyzing the autoregressive, differencing, and moving average components of the series.

[0095] For example, the model predicts that the deviation will increase in the next 5 frames, indicating that the inkjet system may have a persistent deviation.

[0096] Preferably, the parameters of the ARIMA model need to be optimized based on historical data to adapt to different sedimentation rates.

[0097] Specifically, Kalman filtering is used to smooth the deviation sequence and reduce the impact of noise. Assume that the deviation sequence is 5, 6, 4, and 7 square millimeters. After filtering, the smoothed sequence is obtained, such as 5.2, 5.8, 4.3, and 6.5 square millimeters.

[0098] It is understood that the Kalman filter dynamically adjusts the deviation value through state estimation and measurement update to ensure a more stable sequence. The filter parameters need to be calibrated according to the noise characteristics of the inkjet environment.

[0099] For example, if the fluctuation amplitude of the smoothed deviation series exceeds a preset threshold, such as 2 square millimeters, a normal distribution fit is used to determine the mean and standard deviation. Suppose the fitting results show a mean of 5.5 square millimeters and a standard deviation of 0.8 square millimeters, indicating that the deviation distribution is relatively concentrated.

[0100] In one embodiment, the distribution parameters are used to evaluate system stability and guide subsequent parameter adjustments.

[0101] In one embodiment, the gradient descent method updates the proportional, integral, and derivative parameters of the PID controller based on the deviation distribution parameter. Assuming the initial proportional parameter is 0.5, it is iteratively adjusted to 0.55 based on the deviation mean and standard deviation.

[0102] Preferably, the learning rate should be moderate to avoid parameter oscillation. The updated control parameter set improves inkjet accuracy.

[0103] For example, the PID controller calculates the injection frequency based on the updated parameters. Assume that the original frequency is 100 Hz and after adjustment it is 105 Hz, generating a frequency parameter set.

[0104] It should be noted that the frequency adjustment needs to take into account the nozzle response time to ensure a smooth transition.

[0105] Specifically, if the difference between the adjusted frequency parameter set and the historical parameter set exceeds a threshold, such as 5Hz, the root locus method is used to analyze stability. The root locus method verifies that the frequency adjustment will not cause oscillations by examining the location of the system's poles.

[0106] In one embodiment, the analysis results show that the pole is located in a stable region, verifying that the frequency parameter set is valid.

[0107] It is understandable that the final frequency parameter set updates the configuration file in JSON format.

[0108] For example, the configuration file records the frequency of 105 Hz, timestamp and inkjet mode, which facilitates system reading and execution.

[0109] Preferably, the JSON format should have a clear structure to ensure compatibility.

[0110] Step S106 , controlling the deposition process according to the updated jetting frequency profile, and detecting the transition width of the boundary area.

[0111] Image data of the boundary region of the deposition process is obtained, and the boundary region is extracted using an image segmentation algorithm to obtain a segmented boundary image. If the pixel distribution entropy value of the segmented boundary image exceeds a preset entropy threshold, the segmented boundary image is denoised using a median filter to obtain a denoised boundary image. Based on the denoised boundary image, the boundary transition region is extracted using the Canny edge detection algorithm to obtain a boundary transition contour. If the continuity parameter of the boundary transition contour is lower than a preset continuity threshold, the contour breaks are repaired using a morphological dilation operation to obtain a repaired boundary contour. Based on the repaired boundary contour, the pixel width of the boundary transition region is calculated to obtain a boundary transition width sequence. If the variance of the boundary transition width sequence exceeds a preset variance threshold, the width sequence is classified using the K-means clustering algorithm to obtain a classified width sequence. Based on the classified width sequence, the local mean is calculated using a sliding window method to obtain a smoothed width sequence. If the fluctuation amplitude of the smoothed width sequence exceeds a preset amplitude threshold, the smoothed width sequence is optimized using linear interpolation to obtain an optimized width sequence. Based on the optimized width sequence, a boundary transition width profile is generated in JSON format to obtain a width profile. Based on the width profile, control parameters of the deposition process are adjusted to obtain an adjusted control parameter set.

[0112] Specifically, in the inkjet deposition scenario, acquiring image data of the boundary area of the deposition process is a key step.

[0113] For example, an industrial camera captures images of the inkjet area at a frequency of 30 frames per second, generating raw image data including the boundary area, with an image resolution of 1280×720 pixels and a grayscale value of 0 to 255.

[0114] It should be noted that the boundary region generally refers to the interface between the inkjet deposited material and the substrate, and its image characteristics are characterized by a dramatic change in grayscale value. An image segmentation algorithm is used to extract the boundary region and obtain a segmented boundary image.

[0115] Preferably, based on the Otsu threshold segmentation method, by analyzing the image grayscale histogram, the threshold is automatically determined to separate the boundary area from the background.

[0116] For example, assuming the threshold is 150, pixels with grayscale values higher than 150 are classified as boundary areas, generating a binary segmentation image.

[0117] It is understood that the Otsu threshold segmentation method can adapt to image characteristics under different lighting conditions. If the pixel distribution entropy value of the segmented boundary image exceeds the preset entropy value threshold, denoising is performed. The pixel distribution entropy value reflects the degree of disorder in the grayscale distribution of the image pixels.

[0118] In one embodiment, the entropy value is calculated using the Shannon entropy formula. Assuming the calculated result is 5.2 and the preset threshold is 5.0, it indicates that the image is noisy. In this case, a 3×3 window median filter is used to replace each pixel with the median of its neighboring pixels to obtain a denoised boundary image. After denoising, the image edges are smoother and the number of noise points is reduced. Based on the denoised boundary image, the Canny edge detection algorithm is used to extract the boundary transition region. The Canny algorithm generates boundary transition contours through Gaussian filtering, gradient calculation, and double threshold processing.

[0119] For example, a low threshold of 50 and a high threshold of 150 are set to detect continuous boundary lines. If the continuity parameter of the boundary transition contour falls below a preset threshold, such as 80%, the break is repaired using a morphological dilation operation. In one possible implementation, a 3×3 structuring element is used for dilation to fill small gaps in the contour, resulting in a repaired boundary contour. Based on the repaired boundary contour, the pixel width of the boundary transition region is calculated to obtain a boundary transition width sequence.

[0120] Specifically, the pixel width of the boundary area is measured along the normal direction of the contour, and the resulting sequence is assumed to be 10, 12, 9, and 11 pixels. If the sequence variance exceeds a preset threshold, such as 2.0, the K-means clustering algorithm is used to classify the width sequence.

[0121] For example, setting K = 2, the width is divided into two categories: narrower width is about 9 pixels, and wider width is about 12 pixels, generating a classified width sequence. The sliding window method is used to calculate the local mean to obtain a smoothed width sequence.

[0122] In one embodiment, the window size is 3, and the calculated results are 10.3, 10.7, and 10.3 pixels. If the fluctuation amplitude of the smoothed width sequence exceeds a preset threshold, such as 1.0 pixel, the sequence is optimized by linear interpolation.

[0123] For example, we insert intermediate values at points with large fluctuations to generate an optimized width sequence, such as 10.4, 10.6, and 10.4 pixels. Based on the optimized width sequence, we generate a border transition width profile in JSON format.

[0124] Exemplarily, the profile contains a width sequence, a timestamp, and an inkjet pattern, e.g.

[0125] {"widths":[10.4,10.6,10.4],"timestamp":"2025-04-19T10:00:00",

[0126] "mode":"standard"}. Adjust the control parameters of the deposition process according to the width profile.

[0127] For example, based on the width sequence, the nozzle movement speed is adjusted from 50 mm / s to 52 mm / s, and an adjustment control parameter set is generated to improve deposition uniformity.

[0128] In step S107 , if the transition width exceeds the micron-level threshold, an ink deposition dataset including optimized boundary coordinates is generated.

[0129] Acquire a boundary area image, wherein the boundary area image includes pixel distribution characteristics. If the pixel distribution characteristics exceed a preset distribution threshold, the boundary area image is segmented using the Otsu algorithm to obtain a segmented boundary image. Based on the segmented boundary image, the boundary contour is extracted using the Canny edge detection algorithm to obtain an initial boundary contour. If the continuity of the initial boundary contour is lower than a preset continuity threshold, the initial boundary contour is repaired using a closing operation to obtain a repaired boundary contour. Based on the repaired boundary contour, a boundary tracking algorithm is used to calculate a set of coordinate points in the boundary transition area to obtain a boundary coordinate data set. If the coordinate point distribution density of the boundary coordinate data set exceeds a preset density threshold, the boundary coordinate data set is classified using the K-means clustering algorithm to obtain a classified coordinate data set. Based on the classified coordinate data set, cubic spline interpolation is used to optimize the coordinate points in the classified coordinate data set to obtain optimized boundary coordinates. The optimized boundary coordinates are integrated with the ink deposition parameters in JSON format to obtain an ink deposition data set.

[0130] Specifically, in inkjet deposition scenarios, acquiring boundary region images is a key step in analyzing the interface between the deposited material and the substrate. Boundary region images typically contain pixel distribution characteristics, reflecting the spatial variation of grayscale values.

[0131] For example, pixel distribution characteristics can be characterized by calculating the standard deviation of an image's grayscale histogram. For example, for a 1280×720 pixel grayscale image, a standard deviation of 20 and a preset distribution threshold of 15 indicate a high pixel distribution complexity, requiring further processing. The Otsu algorithm automatically determines the segmentation threshold to separate the boundary region from the background by analyzing the grayscale histogram.

[0132] For example, the threshold is 140, and pixels with grayscale values higher than 140 are classified as boundary areas, generating a binary segmentation image.

[0133] It's understandable that the Otsu algorithm can adapt to varying lighting conditions, improving segmentation robustness. Based on the segmented boundary image, the Canny edge detection algorithm is used to extract boundary contours. The Canny algorithm smoothes the image using a Gaussian filter and then calculates the gradient to detect edge points.

[0134] Preferably, the initial boundary contour is generated by setting the low threshold to 60 and the high threshold to 160. If the continuity of the initial boundary contour is lower than a preset threshold, such as 85%, it needs to be repaired.

[0135] In one embodiment, a closing operation fills small breaks in the contour by dilating and then eroding a 3×3 structuring element, resulting in a repaired boundary contour. This closing operation effectively connects discontinuous edges and enhances contour integrity. Based on the repaired boundary contour, a boundary tracing algorithm calculates the coordinate point set for the boundary transition region.

[0136] It should be noted that the boundary tracing algorithm traverses the contour pixel by pixel and records the coordinates of each boundary point.

[0137] For example, if tracking generates a boundary coordinate dataset containing 5,000 coordinate points, classification processing is required if the distribution density of coordinate points exceeds a preset threshold, such as the number of coordinate points per 10 pixels is greater than 5.

[0138] Specifically, the K-means clustering algorithm divides coordinate points into two categories, such as dense areas and sparse areas. Assuming that the clustering results are 3,000 dense points and 2,000 sparse points, a classified coordinate dataset is generated. Clustering helps distinguish the complexity and regularity of boundary areas. Cubic spline interpolation is used to optimize the coordinate points in the classified coordinate dataset. Cubic spline interpolation generates a smooth curve by fitting the coordinate points with a piecewise polynomial. In one possible implementation, the spacing between coordinate points is uniformed after interpolation, and the optimized boundary coordinates are more consistent with the actual boundary curve shape. The smoothed coordinate point set can improve the accuracy of subsequent analysis. The optimized boundary coordinates and ink deposition parameters are integrated in JSON format to generate an ink deposition dataset.

[0139] For example, a dataset contains a sequence of boundary coordinates, an inkjet flow rate such as 0.5 ml / s, and a timestamp such as 2025-04-19T12:00:00, in the format of {"coordinates":"[(x1,y1),

[0140] (x2, y2), ...]", "flow": "0.5", "timestamp": "2025-04-19T12:00:00"}. The JSON structure facilitates parameter passing and storage, which helps to accurately control the deposition process.

[0141] In one embodiment, the boundary coordinate data set may be further combined with the nozzle position parameters to generate a dynamic adjustment solution.

[0142] For example, based on the coordinate points of dense areas, the nozzle path is adjusted to reduce deposition overlap.

[0143] Preferably, the inkjet frequency can be increased in sparse areas to ensure uniform coverage. These adjustment schemes improve deposition quality and consistency through the coordinated optimization of multi-dimensional parameters.

[0144] Step S108 : generating zone heating control parameters corresponding to the zone identifiers and curing temperatures based on the ink deposition dataset.

[0145] An ink deposition dataset is obtained, comprising boundary coordinates and ink deposition parameters. Gaussian filtering is used to denoise the ink deposition parameters to obtain a preprocessed deposition parameter set. Region identifiers are extracted using threshold segmentation based on the gradient distribution of the preprocessed deposition parameter set. The region identifiers are partitioned using a quadtree spatial partitioning algorithm, with the minimum partition area set to a preset value to obtain a partition identifier set. The coefficient of variation of the boundary points of the partition identifier set is calculated. If the coefficient of variation is greater than a preset threshold, region merging is performed. Adjacent partitions are detected based on Delaunay triangulation. If the standard deviation of the area of adjacent partitions is greater than a preset threshold, they are merged into the same partition to obtain an adjusted partition identifier set. For the adjusted partition identifier set, the mean pixel coordinates of each partition are calculated to obtain a geometric center coordinate set. Based on the geometric center coordinate set and a preset curing temperature lookup table, a two-dimensional bilinear interpolation algorithm is used to generate partition heating control parameters to obtain a partition heating control parameter set. The heating parameter value of each partition is extracted from the partition heating control parameter set and a sliding window mean filter is applied to obtain a smoothed heating parameter set. The range of the smoothed heating parameter set is calculated. If the range exceeds a preset threshold, the parameters of adjacent partitions are weighted averaged using the inverse distance weighting method, with the weight coefficient being the inverse of the partition center distance, to obtain the optimized heating parameter set. Based on the optimized heating parameter set, a linear proportional conversion is used to generate PWM duty cycle instructions, resulting in the partition heating control instruction set.

[0146] It should be noted that the aforementioned Delaunay triangulation constructs a triangular network that satisfies certain properties given a set of points. This network divides all points into triangles, aims to maximize the minimum angle of each triangle, and ensures that the circumcircle of each triangle contains no other points. Therefore, this embodiment uses Delaunay triangulation to detect adjacent partitions, which can improve computational stability and optimize computational efficiency. It can also simplify adjacent partition detection and adapt to dynamic changes.

[0147] Specifically, obtaining ink deposition datasets is a core step in the inkjet printing process. The datasets usually contain boundary coordinates and ink deposition parameters such as nozzle flow or deposition time.

[0148] For example, boundary coordinates can be obtained through image processing, while ink deposition parameters are extracted from inkjet device logs, such as a nozzle flow rate of 0.3 ml / s and a deposition time of 2 seconds. This dataset provides the basis for subsequent optimization. Gaussian filtering is used to reduce noise in the ink deposition parameters.

[0149] It should be noted that Gaussian filtering reduces noise interference by weighted averaging smoothing parameters.

[0150] In one embodiment, for nozzle flow data, the Gaussian kernel standard deviation is set to 1.5, and the flow fluctuation after filtering is reduced from ±0.05 ml / s to ±0.01 ml / s to ensure parameter stability. Based on the gradient distribution of the pre-processed deposition parameter set, the region identifier is extracted through threshold segmentation.

[0151] Specifically, the gradient reflects the spatial rate of change of the parameter, and high-gradient areas usually correspond to heterogeneous deposition.

[0152] Preferably, the gradient threshold is set to 0.1, and the high-variation region is segmented to generate an identification image, where an identification value of 1 indicates a region to be optimized. A quadtree space partitioning algorithm is used to partition the region identification.

[0153] As you can understand, the quadtree recursively partitions a region into subregions of varying sizes. In one possible implementation, the minimum partition size is set to 100 pixels, generating a partition identifier set containing 50 subregions to facilitate refined analysis. The coefficient of variation of the boundary points of the partition identifier set is calculated, and if it exceeds a preset threshold, the region is merged.

[0154] For example, the coefficient of variation is 0.4, which exceeds the threshold of 0.3, indicating that the boundary points are unevenly distributed.

[0155] In one embodiment, adjacent partitions with fewer than 10 boundary points are merged to reduce fragmented partitions and improve processing efficiency. Adjacent partitions are detected and merged based on Delaunay triangulation.

[0156] For example, triangulation constructs the topological relationships between partitions. If the standard deviation of the area of adjacent partitions is greater than 200 pixels, they are merged into a single partition. After adjustment, the number of partitions is reduced from 50 to 30, and an optimized partition identifier set is generated to reduce computational complexity. For this adjusted partition identifier set, the mean pixel coordinates of each partition are calculated to obtain the geometric center coordinate set.

[0157] Specifically, a partition contains 1000 pixels with a mean coordinate of (500, 300), forming a set of 30 center points that represent the spatial distribution of the partition. Based on the geometric center coordinate set and the curing temperature lookup table, two-dimensional bilinear interpolation is used to generate the partition heating control parameters.

[0158] Preferably, the lookup table defines a temperature range of 100-150 degrees Celsius. After interpolation, the heating temperature of a certain zone is 120 degrees Celsius, generating a parameter set containing 30 temperature values to ensure accurate zone heating. The heating parameter values are extracted from the zone heating control parameter set and smoothed using a sliding window mean filter.

[0159] In one embodiment, the window size is 3, and after filtering, the temperature fluctuation is reduced from ±5 degrees Celsius to ±2 degrees Celsius, generating a smoothed heating parameter set to improve control stability. The range of the smoothed heating parameter set is calculated, and if it is greater than a preset threshold, the inverse distance weighted optimization method is used.

[0160] For example, a temperature range of 20 degrees Celsius exceeds a threshold of 15 degrees Celsius. In one possible implementation, the reciprocal of the partition center distance is used as the weight, and after weighted averaging, the temperature difference is reduced to 10 degrees Celsius, generating an optimized heating parameter set. Based on this optimized heating parameter set, a linear proportional conversion is used to generate a PWM duty cycle instruction.

[0161] For example, a temperature of 120 degrees Celsius corresponds to a duty cycle of 60%, and a partition heating control instruction set containing 30 instructions is generated and directly used for equipment control to ensure heating uniformity.

[0162] Step S109 , obtaining actual temperature distribution data during the curing process, and generating a new power control instruction if the temperature deviation exceeds a threshold.

[0163] The temperature distribution data during the curing process is obtained from the temperature sensor to obtain the real-time temperature distribution. The temperature deviation is calculated based on the comparison between the real-time temperature distribution and the preset standard temperature distribution. The temperature deviation is obtained by subtracting the preset standard temperature distribution from the real-time temperature distribution. If the temperature deviation exceeds the preset threshold, the adjustment amount is calculated using the temperature deviation and the PID controller to generate a power control instruction. The power control instruction is executed by the control system to adjust the power output of the curing equipment to obtain an updated temperature distribution. The updated temperature distribution data is obtained from the temperature sensor and compared with the preset standard temperature distribution to determine whether the temperature deviation is within the preset threshold. If the temperature deviation is still outside the preset threshold, the power control parameter is adjusted using the gradient descent method to generate a new power control instruction.

[0164] Specifically, obtaining temperature distribution data during the curing process from a temperature sensor is a key step in the inkjet printing curing process.

[0165] Exemplarily, a temperature sensor array is arranged above the curing area to collect the temperature value of each pixel in real time to form a two-dimensional temperature distribution map.

[0166] It should be noted that the sensor usually uses infrared temperature measurement technology with an accuracy of up to ±1 degree Celsius.

[0167] For example, in a curing area measuring 1000 x 1000 mm, the sensor collects temperature data at 10,000 points at 10 mm intervals, generating a temperature distribution map with a maximum temperature of 150°C and a minimum temperature of 100°C. This high-resolution data provides the basis for subsequent deviation analysis. The temperature deviation is calculated by comparing the real-time temperature distribution with a preset standard temperature distribution.

[0168] Specifically, the preset standard temperature distribution is a temperature template designed based on an ideal curing effect, for example, the target temperature is 120 degrees Celsius, and a fluctuation of ±5 degrees Celsius is allowed.

[0169] In one embodiment, the real-time temperature distribution shows that the temperature of a certain area is 130 degrees Celsius, and the deviation is 130-120=10 degrees Celsius.

[0170] Preferably, the deviation calculation covers the entire curing area, generating a deviation distribution map that visually reflects areas of temperature unevenness. This deviation analysis helps pinpoint areas requiring adjustment. If the temperature deviation exceeds a preset threshold, a PID controller calculates the adjustment amount and generates a power control command.

[0171] It can be understood that the PID controller adjusts the output power through the proportional, integral and differential parts and responds quickly to deviations.

[0172] For example, the preset threshold is ±5 degrees Celsius, and a deviation of 10 degrees Celsius in a certain area exceeds the threshold.

[0173] In one possible implementation, a PID controller adjusts heater power based on the deviation, increasing the output power control command from 50% to 60% to reduce the temperature deviation. This approach ensures precise and responsive power adjustments. The control system executes the power control command, adjusting the curing equipment's power output and generating an updated temperature profile.

[0174] Specifically, the control system converts the command into an actual power signal for the heater, for example, converting a 60% power command into a 300-watt output.

[0175] In one embodiment, after adjustment, the temperature of a certain area drops from 130 degrees Celsius to 122 degrees Celsius, and the deviation is reduced to 2 degrees Celsius.

[0176] It should be noted that the updated temperature distribution is collected again by the sensor to ensure real-time data. This closed-loop control method improves the stability of temperature regulation. The updated temperature distribution data is obtained from the temperature sensor and compared with the preset standard temperature distribution to determine whether the temperature deviation is within the preset threshold.

[0177] For example, after the update, the deviation in a certain area is 2 degrees Celsius, which falls within the ±5 degrees Celsius threshold range, indicating that the adjustment is effective.

[0178] In one possible implementation, if all regional deviations meet the threshold, the system maintains the current power output; otherwise, it proceeds to the next optimization step. This deviation determination mechanism ensures temperature consistency throughout the curing process. If the temperature deviation remains outside the preset threshold, the power control parameters are adjusted using a gradient descent method to generate new power control instructions.

[0179] Preferably, the gradient descent method iteratively optimizes the controller parameters to gradually reduce the deviation.

[0180] For example, the deviation in a certain area is still 8 degrees Celsius.

[0181] In one embodiment, the gradient descent method adjusts the proportional coefficient in the PID parameters from 1.0 to 0.8, generating a new power command that fine-tunes the power from 60% to 58%. This method avoids overshoot through small-step optimization, ensuring that the temperature distribution gradually approaches the target.

[0182] Step S1010 , measuring the final conductive performance parameters and generating a verification data set corresponding to the region identifiers and the measured conductivity.

[0183] Obtain a raw data set, the raw data set comprising multiple records, each record containing a region identifier and measured conductivity data. For missing conductivity values in the raw data set, calculate the conductivity mean for each group by region identifier, and interpolate missing values using the within-group mean to obtain a complete conductivity data set. Perform Z-score normalization on the conductivity columns of the complete conductivity data set to obtain a standardized conductivity data set. Analyze the standardized conductivity data set using the elbow method to determine the optimal number of clusters, thereby obtaining the number of clusters. Cluster the standardized conductivity data set using the K-means algorithm to obtain a conductivity data set with category labels. Based on the conductivity data set with category labels, calculate the conductivity mean and standard deviation for each cluster by category identifier to obtain statistical characteristics for each cluster. If the standard deviation of a cluster is greater than a preset threshold, mark the cluster as an abnormal category, thereby obtaining an abnormal category set. Extract the region identifiers corresponding to the abnormal category set from the complete conductivity data set to obtain an abnormal region set. According to the region identifiers in the abnormal region set, the measured conductivity records are indexed and matched from the original data set to obtain an abnormal conductivity verification set.

[0184] Specifically, obtaining area identification and measured conductivity data through sensors and storing them as raw data sets is an important step in monitoring material properties in the inkjet printing curing process.

[0185] Exemplarily, the sensor array is arranged in the curing area to collect the conductivity value and corresponding identification of each area.

[0186] For example, the curing area size is 500×500 mm, and the sensor collects data at 10,000 points at 5 mm intervals. At each point, the area identifier such as A1, B2, etc. and the conductivity value such as 10mS / cm are recorded.

[0187] It should be noted that conductivity reflects the material's electrical conductivity and directly affects curing quality. Sensor accuracy is typically ±0.1 mS / cm. The collected data is stored in CSV format, containing a region identifier column and a conductivity column, forming the raw data set. For missing conductivity values in the raw data, we interpolate by calculating the mean of each group, grouped by region identifier.

[0188] In one possible implementation, a region A1 contains 100 points, of which 10 have missing conductivity values. The average conductivity value of the 90 valid points in region A1 is calculated to be 12 mS / cm, and this value is used to fill in the missing points.

[0189] Preferably, before interpolation, check the data distribution to ensure representativeness of the mean. After interpolation, generate a complete conductivity dataset, preserving the region identifiers and conductivity columns to ensure data integrity. Perform Z-score normalization on the conductivity columns of the complete conductivity dataset to eliminate dimensionality effects.

[0190] It can be understood that the Z-score converts the conductivity value into a standard value with a mean of 0 and a standard deviation of 1, which is convenient for subsequent clustering.

[0191] For example, if the conductivity at a point is 15 mS / cm, the mean of the dataset is 12 mS / cm, the standard deviation is 2 mS / cm, and the normalized value is 1.5. This normalized dataset is more suitable for machine learning algorithms. The elbow method is used to determine the optimal number of clusters, k, and the K-means algorithm is used for clustering.

[0192] In one embodiment, the elbow method finds the inflection point k = 3 by plotting the number of clusters against the sum of squared errors within the clusters. The K-means algorithm divides the normalized data into three clusters and outputs a dataset with class labels, where each row records the region identifier, conductivity, and cluster label (e.g., C1, C2).

[0193] It should be noted that the K-means algorithm iteratively optimizes cluster centers to classify regions with similar conductivity. Based on the clustering results, the mean and standard deviation of the conductivity of each cluster are calculated by grouping by category label.

[0194] Specifically, cluster C1 contains 5,000 points with a mean of 11 mS / cm and a standard deviation of 0.3. Cluster C2 has a mean of 13 mS / cm and a standard deviation of 0.6. The preset standard deviation threshold is 0.5. Cluster C2 is marked as an anomaly if its standard deviation exceeds the threshold.

[0195] Preferably, a high standard deviation reflects large fluctuations in conductivity, which may indicate uneven solidification. Region identifiers corresponding to abnormal categories are extracted from the complete conductivity data set to form an abnormal region set.

[0196] For example, cluster C2 contains 1,000 identifiers for regions B1 and B2, forming a set of abnormal regions. Based on this set, the measured conductivity records are indexed and matched from the original dataset to generate an abnormal conductivity verification set.

[0197] In one embodiment, the validation set includes conductivity records of the B1 region, such as 13.2 mS / cm and 14.1 mS / cm, for subsequent analysis of abnormal causes.

[0198] It is understandable that the validation set provides accurate data support for process optimization.

[0199] It will be apparent to those skilled in the art that the present application is not limited to the details of the exemplary embodiments described above and that the present application can be implemented in other specific forms without departing from the spirit or essential characteristics of the present application. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the present application is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. A method for preparing a PCB board with high carbon ink, characterized in that: The method comprises: Obtaining a conductive property distribution data set including region identification, conductive property values, and boundary coordinates; Generating a mapping relationship between nozzle number, ink type and target area coordinates according to the conductive performance distribution data set; A micron-level precision control algorithm is used to generate motion control instruction sets corresponding to timestamps, nozzle positions, and deposition coordinates; Acquire an ink deposition image data set during the deposition process and calculate a deviation between an actual deposition amount and a target deposition amount; If the deviation value exceeds a preset threshold, updating an injection frequency profile including proportional-integral-derivative parameters; controlling the deposition process based on the updated jet frequency profile and detecting the transition width in the boundary region; If the transition width exceeds a micrometer threshold, generating an ink deposition dataset comprising optimized boundary coordinates; generating a zone heating control parameter corresponding to a zone identifier and a curing temperature based on the ink deposition data set; Acquire actual temperature distribution data during the curing process and generate new power control instructions if the temperature deviation exceeds a threshold; Measure the final conductive performance parameters and generate a validation dataset with region identification corresponding to the measured conductivity.

2. The method according to claim 1, characterized in that The obtaining of a conductive property distribution data set including region identifiers, conductive property values, and boundary coordinates includes: Acquire an initial data set from a sensor network, the initial data set comprising region identifiers, conductivity values, and boundary coordinates; Performing noise detection on the initial data set to obtain a denoised data set; Based on the denoised data set, a K-means clustering algorithm is used to divide the regions to obtain a region division result; extracting the conductivity value of each region from the denoised data set to generate a property distribution feature; Calculating the variance of the performance distribution characteristics, and if the variance is greater than a preset threshold, smoothing the conductivity value using a linear interpolation algorithm to obtain a smoothed performance distribution; generating a conductive performance distribution data set including region identifiers, performance values, and boundary coordinates according to the smoothed performance distribution and the region division result; The boundary coordinates of the conductive property distribution dataset and the initial dataset are compared by a boundary coordinate consistency verification algorithm to determine a final conductive property distribution dataset.

3. The method according to claim 1, characterized in that Generating a mapping relationship between a nozzle number, an ink type, and target area coordinates according to the conductive performance distribution data set includes: Based on the conductive performance distribution data set, an initial data table is generated using a data extraction algorithm; Divide the regions using a K-means clustering algorithm according to the region identifiers and boundary coordinate points in the initial data table to obtain a region division result; If the boundary coordinate points in the region division result are inconsistent with the boundary coordinate points in the initial data table, then using a coordinate consistency verification algorithm to adjust the region division result to obtain consistent region division; Extracting a region identifier and target region coordinates of each region from the consistent region division, and generating a region coordinate table including the region identifier and the target region coordinates; According to the region identifier in the region coordinate table, a preset nozzle allocation rule is used to match the nozzle number value and the ink type name to obtain a nozzle allocation table; Extract the nozzle number value, ink type name and target area coordinates from the nozzle allocation table to obtain a mapping relationship table; According to the target area coordinates in the mapping relationship table and the boundary coordinate points in the consistent area division, the mapping relationship table is verified using a coordinate consistency verification algorithm to obtain a final mapping relationship table.

4. The method according to claim 1, wherein The micron-level precision control algorithm is used to generate a motion control instruction set corresponding to a timestamp, nozzle position, and deposition coordinates, including: Obtaining a timestamp and nozzle motion trajectory data from the time series data, and using a threshold segmentation method to separate the position coordinates in the nozzle motion trajectory data to obtain a nozzle position coordinate set; Obtain the target deposition coordinates through the visual sensor to obtain the initial coordinate data set; According to the timestamp and nozzle position coordinates in the initial coordinate data set, the motion acceleration of adjacent coordinate points is calculated using Kalman filtering. If the motion acceleration exceeds a preset acceleration threshold, it is determined that there is a discontinuity in the nozzle motion trajectory; Generating new coordinate points between the discontinuity points using cubic spline interpolation to obtain a smooth motion trajectory, wherein the velocity change rate of the smooth motion trajectory is lower than a preset smoothness threshold; Extracting the interpolated nozzle position coordinates from the smooth motion trajectory, establishing a mapping relationship with the target deposition coordinates through time window matching, and obtaining a mapping relationship table between the timestamp, the nozzle position coordinates, and the target deposition coordinates; According to the target deposition coordinates corresponding to each timestamp in the mapping relationship table, a preset instruction template is matched to generate a preliminary instruction set; The actual motion path when the preliminary instruction set is executed is measured by a laser interferometer. If the deviation between the actual motion path and the target deposition coordinate exceeds a preset tolerance threshold, the parameters of the preliminary instruction set are adjusted to obtain a final motion control instruction set.

5. The method according to claim 1, characterized in that The step of obtaining an ink deposition image dataset during the deposition process and calculating a deviation between an actual deposition amount and a target deposition amount includes: Collecting images of the ink deposition process and intercepting them at preset time intervals to generate an original image set; Processing the original image set using an image segmentation algorithm to extract the outline of the ink deposition area to generate a segmented image set; Calculating the pixel area of each contour region in the segmented image set using a contour area calculation function to generate an actual sedimentation dataset; Read the target sedimentation value from the configuration file to generate a target sedimentation data set; If the absolute difference between the actual deposition amount and the target deposition amount exceeds a preset threshold, a mean filter function is used to filter the difference sequence to generate a deposition deviation data set; Matching the deposition deviation data set with a pre-established compensation parameter comparison table to generate an adjustment parameter set; the compensation parameter comparison table includes a correspondence between the deviation range and the compensation coefficient; According to the compensation coefficient in the adjustment parameter set, a contour detection function is used to re-detect the boundaries of the deposition area in the segmented image set to generate an optimized deposition amount data set.

6. The method according to claim 1, characterized in that If the deviation value exceeds a preset threshold, updating the injection frequency configuration file including proportional-integral-derivative parameters includes: Obtaining a sedimentation deviation value sequence, and if the sedimentation deviation value sequence exceeds a preset threshold, extracting a time series feature of the sedimentation deviation value sequence through an ARIMA model to obtain a deviation change trend; According to the deviation change trend, the deposition deviation value sequence is smoothed by using Kalman filtering to obtain a smoothed deviation sequence; If the fluctuation amplitude of the smoothed deviation sequence exceeds a preset amplitude threshold, the normal distribution is used to fit the smoothed deviation sequence to determine the deviation distribution parameters; According to the deviation distribution parameters, a gradient descent method is used to update the proportional parameters, integral parameters and differential parameters in the configuration file to obtain an updated control parameter set; Calculating the injection frequency value according to the updated control parameter set by a PID controller to obtain an adjusted frequency parameter set; If the difference between the adjusted frequency parameter set and the historical frequency parameter set exceeds a preset difference threshold, a root locus method is used to perform stability analysis on the adjusted frequency parameter set to obtain a verified frequency parameter set; The injection frequency configuration file is updated according to the verified frequency parameter set to obtain a final frequency configuration file.

7. The method according to claim 1, characterized in that The method of controlling the deposition process according to the updated jetting frequency profile and detecting the transition width of the boundary area includes: Acquire image data of the boundary area of the deposition process, extract the boundary area using an image segmentation algorithm, and obtain a segmented boundary image; If the pixel distribution entropy value of the segmented boundary image exceeds a preset entropy value threshold, the segmented boundary image is subjected to denoising processing using a median filter to obtain a denoised boundary image; According to the denoised boundary image, a Canny edge detection algorithm is used to extract the boundary transition area to obtain a boundary transition contour; If the continuity parameter of the boundary transition contour is lower than a preset continuity threshold, repairing the contour break by a morphological dilation operation to obtain a repaired boundary contour; Calculating the pixel width of the boundary transition area according to the repaired boundary contour to obtain a boundary transition width sequence; If the variance of the boundary transition width sequence exceeds a preset variance threshold, the width sequence is classified using a K-means clustering algorithm to obtain a classified width sequence; According to the classification width sequence, a sliding window method is used to calculate the local mean to obtain a smooth width sequence; If the fluctuation amplitude of the smoothed width sequence exceeds a preset amplitude threshold, the smoothed width sequence is optimized by linear interpolation to obtain an optimized width sequence; generating a boundary transition width profile according to the optimized width sequence to obtain a width profile; The control parameters of the deposition process are adjusted according to the width profile to obtain an adjusted control parameter set.

8. The method according to claim 1, characterized in that If the transition width exceeds a micron-level threshold, generating an ink deposition dataset containing optimized boundary coordinates, comprising: Acquire a boundary region image, wherein the boundary region image includes pixel distribution characteristics; If the pixel distribution characteristic exceeds a preset distribution threshold, segmenting the boundary area image using an Otsu algorithm to obtain a segmented boundary image; Extracting the boundary contour using the Canny edge detection algorithm based on the segmented boundary image to obtain an initial boundary contour; If the continuity of the initial boundary contour is lower than a preset continuity threshold, repairing the initial boundary contour by a closing operation to obtain a repaired boundary contour; According to the repaired boundary contour, a boundary tracking algorithm is used to calculate a coordinate point set of a boundary transition area to obtain a boundary coordinate data set; If the distribution density of the coordinate points of the boundary coordinate data set exceeds a preset density threshold, classifying the boundary coordinate data set using a K-means clustering algorithm to obtain a classified coordinate data set; According to the classification coordinate data set, the coordinate points in the classification coordinate data set are optimized using cubic spline interpolation to obtain optimized boundary coordinates; The optimized boundary coordinates are integrated with ink deposition parameters to obtain an ink deposition data set.

9. The method according to claim 1, characterized in that The generating of the zone identification and the zone heating control parameter corresponding to the curing temperature based on the ink deposition data set includes: Acquire an ink deposition dataset, wherein the ink deposition dataset includes boundary coordinates and ink deposition parameters; Using Gaussian filtering to perform noise reduction processing on the ink deposition parameters to obtain a pre-processed deposition parameter set; Extracting region identification by threshold segmentation according to the gradient distribution of the pre-processing deposition parameter set; Partitioning the region identifiers using a quadtree space partitioning algorithm, setting a minimum partition area as a preset value, and obtaining a partition identifier set; Calculating the coefficient of variation of the boundary points of the partition identification set, and performing region merging if the coefficient of variation is greater than a preset threshold; Adjacent partitions are detected based on Delaunay triangulation. If the standard deviation of the area of adjacent partitions is greater than a preset threshold, they are merged into the same partition to obtain an adjusted partition identification set. Calculating the mean of pixel coordinates of each partition for the adjusted partition identifier set to obtain a geometric center coordinate set; According to the geometric center coordinate set and the preset curing temperature lookup table, a two-dimensional bilinear interpolation algorithm is used to generate the zone heating control parameters to obtain a zone heating control parameter set; Extracting a heating parameter value for each partition from the partition heating control parameter set, and applying a sliding window mean filter to obtain a smoothed heating parameter set; Calculating the range of the smoothed heating parameter set; if the range is greater than a preset threshold, performing weighted averaging on adjacent partition parameters using an inverse distance weighted method, where the weight coefficient is the inverse of the partition center distance, to obtain an optimized heating parameter set; According to the optimized heating parameter set, a PWM duty cycle instruction is generated by using linear proportional conversion to obtain a partition heating control instruction set.

10. The method according to claim 1, characterized in that The actual temperature distribution data during the curing process is obtained, and if the temperature deviation exceeds a threshold, a new power control instruction is generated, including: Acquire temperature distribution data during the curing process from the temperature sensor to obtain real-time temperature distribution; Comparing the real-time temperature distribution with the preset standard temperature distribution and calculating the temperature deviation, wherein the temperature deviation is obtained by subtracting the preset standard temperature distribution from the real-time temperature distribution; If the temperature deviation exceeds a preset threshold, the temperature deviation is used with a PID controller to calculate an adjustment amount and generate a power control instruction; The power control instruction is executed by the control system to adjust the power output of the curing equipment to obtain an updated temperature distribution; Acquire the updated temperature distribution data from the temperature sensor, compare it with a preset standard temperature distribution, and determine whether the temperature deviation is within a preset threshold; If the temperature deviation is still outside the preset threshold, the power control parameter is adjusted using a gradient descent method to generate a new power control instruction.

Images