A method for fast search and match of jetting patterned orifice-print dot

Abstract

The present application belongs to the technical field of pattern printing and thin film packaging, and particularly relates to a method for quickly searching and matching of printing points of a patterned printing nozzle, comprising: determining a maximum continuous nozzle interval of a print head in a printing process according to the spatial adjacency relationship of actual jetting landing points of each nozzle; continuously partitioning a new printing dot array in the Y direction to determine a printing dot partition in which each nozzle can participate in printing; determining a printing point candidate space of each nozzle based on the Y direction distance between the actual jetting landing point corresponding to each nozzle and each printing point that can participate in printing; determining the priority weight of each printing point in the printing point candidate space to use the nozzle for printing according to the current jetting state of each nozzle; modeling the matching process between the nozzle of the print head and the printing point as a constrained maximum weight matching optimization problem to determine the matching relationship between the nozzle and the printing point by solving. The present application can improve the search and matching efficiency in the patterned printing process of thin film packaging and improve the printing effect.

Description

Technical Field

[0001] This invention belongs to the technical field of patterned printing and thin film encapsulation, and more specifically, relates to a method for rapid search and matching of nozzles and printing dots in patterned inkjet printing. Background Technology

[0002] With the development of high-resolution displays, flexible electronics, and functional thin-film devices, patterned printing technologies based on piezoelectric and electrofluid jetting processes are widely used in the precise deposition of thin film encapsulation (TFE) at the micro-nano scale. In this type of patterned printing process, it is typically necessary to establish a one-to-one matching relationship between large-sized ink droplet arrays and multiple nozzles in the printhead to achieve precise control of the jetting position and dosage, thereby meeting the requirements for film deposition accuracy and pattern consistency.

[0003] In existing technologies, the search methods for determining whether there is a feasible correspondence between nozzles and printing points often employ global traversal or binary search to retrieve the relationship between nozzles and printing points. However, as the pattern size and resolution continue to increase, the number of nozzles and printing points increases significantly, and the search space for candidate relationships expands rapidly, resulting in low search efficiency and high computational overhead, making it difficult to meet the real-time and engineering efficiency requirements of large-size patterned printing.

[0004] After obtaining the correspondence between nozzles and print points, it is still necessary to further determine the final matching result. In existing technologies, point-by-point search and greedy allocation methods are often used. Without fully considering the global constraints of nozzle resources and the competition between print points, in large-size patterned printing scenarios, the number of nozzles and the number of print points often present a high-dimensional, many-to-many complex relationship. This type of greedy matching method easily ignores the global constraints and local competition of nozzle resources, resulting in multiple print points competing for the same nozzle or uneven utilization of nozzle resources, thereby causing local conflicts and affecting the final printing effect. Summary of the Invention

[0005] In view of the above-mentioned defects or improvement needs of the prior art, the present invention provides a rapid search and matching method for nozzle-printing dots in inkjet patterning, which aims to improve the search and matching efficiency and improve the printing effect in the thin film encapsulation patterning printing process.

[0006] To achieve the above objectives, according to one aspect of the present invention, a method for rapid search and matching of nozzles and print dots for patterned printing is provided, comprising: Based on the spatial adjacency of the actual spray landing points of each nozzle in the array, the distance between adjacent actual spray landing points is calculated. If the distance exceeds a preset break threshold, it is determined that there is a break at that distance position, thereby dividing the array of actual spray landing points of all nozzles into multiple continuous intervals. Among the multiple continuous intervals, the longest interval is selected as the maximum printable continuous nozzle interval during the printing process. All print points in the preset print dot matrix, except for the first row, are mapped to the first row along a direction perpendicular to the Y direction to obtain a new print dot matrix. The new print dot matrix is ​​then continuously partitioned in the Y direction, with overlapping partitions. Each nozzle in the largest printable continuous nozzle interval is traversed. Taking the first print point in the Y direction of the new print dot matrix as the zero point coordinate in the Y direction, the ratio of the Y-coordinate value of the actual spray landing point corresponding to the nozzle to the minimum distance between adjacent points in the new print dot matrix is ​​calculated. Based on this ratio, the nearest neighbor print point to the landing point is found, thereby determining the print dot partition in which the nozzle can participate in printing. The Y-distance between the actual spray landing point corresponding to the nozzle and each print point in the print dot partition in the print dot partition in the print dot matrix is ​​calculated. The set of print points whose Y-distance is not greater than a preset threshold is used as the candidate space of print points for the nozzle. Based on the Y-axis positional deviation between the actual spray landing point of each nozzle and each print point in the candidate space of the nozzle print point, and the current spraying state of the nozzle, the priority weight for each print point in the candidate space of the nozzle to be printed by that nozzle is determined. The matching process between the nozzle and the print point is modeled as a constrained maximum weight matching optimization problem. The objective is to maximize the overall matching weight so that nozzles with smaller spatial deviations and better spraying states are selected first, thereby obtaining the overall optimal nozzle-print point matching relationship. The constraints include that each print point is printed by one nozzle and each nozzle prints at most one print point. The matching relationship between the nozzle and the print point is determined by solving the maximum weight matching optimization problem.

[0007] Furthermore, the method also includes: Identify the matching relationship between nozzles and printing points. If the same nozzle is selected by multiple printing points in all matching relationships, determine all nozzles that can participate in printing at each of the multiple printing points. Based on the priority weight of each printing point selecting the corresponding nozzle for printing, construct a corresponding local nozzle-printing point matching problem. The goal of this problem is to maximize the matching weight of the local area, so that nozzles with smaller spatial deviations and better spraying states are selected first, thereby obtaining the locally optimal nozzle-printing point matching relationship. Under the condition of satisfying the constraints, perform local maximum weight matching optimization to solve the problem and determine the final matching relationship between nozzles and printing points in the local area.

[0008] Furthermore, the disconnection threshold is determined based on the target film-forming spacing, the actual jetting point Y-axis deviation, and the allowable film-forming continuity deviation.

[0009] Furthermore, the preset print dot matrix is ​​a uniformly spaced print dot matrix, and the continuous partitioning method is as follows: Number the printed dots on the new printed dot matrix along the Y direction, and denote them as follows: , n is the total number of print dots on the new print dot matrix, and the spacing between adjacent print dots in the new print dot matrix is ​​denoted as n. Based on the numbering order of the print dots, the new print dot matrix is ​​divided into multiple local print dot partitions, each partition being denoted as... Define it as ; The method for determining the print dot partitioning that a nozzle can participate in printing based on the ratio is as follows: The ratio corresponding to each nozzle is rounded down and expressed as follows: In the formula, This indicates the nozzle The corresponding Y-coordinate value of the actual spray landing point is used to determine the nozzle. The actual spray point is mapped to the partition. That is, the nozzle The print dots that can participate in printing are divided into zones. .

[0010] Furthermore, the preset threshold is determined based on the minimum spacing between adjacent print dots in the new print dot matrix and the spreading performance of the ink to be printed.

[0011] Furthermore, the current spraying state of each nozzle is determined in the following way: Obtain the jetting deviation sequence of the nozzle in the first t printing cycles, wherein the jetting deviation sequence includes the Y-direction position deviation between the actual jetting point of the nozzle and the target position, the jetting volume deviation and / or other temporal indicators that affect the jetting state; The jetting deviation sequence is input into a pre-trained Long Short-Term Memory (LSTM) network to predict the possible current jetting state of the nozzle in subsequent printing.

[0012] Furthermore, the objective function of the maximum weight matching optimization problem is: In the formula, Used to indicate nozzle With print point The matching relationship between them, when the nozzle With print point During matching, ,otherwise ; Indicates the print point Select spray nozzle Print priority matches weight.

[0013] According to another aspect of the present invention, a patterned inkjet printing system is provided, characterized in that it is equipped with an inkjet planning module, wherein the inkjet planning module uses the nozzle-printing point fast search and matching method described above to realize the correspondence between the printing point and the nozzle in the Y-direction spatial position.

[0014] According to another aspect of the invention, a computer-readable storage medium is provided, the computer-readable storage medium including a stored computer program, wherein, when the computer program is run by a processor, it controls the device where the storage medium is located to perform the steps of the method described above.

[0015] In summary, compared with the prior art, the technical solutions conceived by this invention have the following main advantages: 1. This invention proposes a fast search and matching method for nozzle-printing dots in patterned printing. The method first determines the maximum printable continuous interval of the nozzle during the printing process. Under the constraint of this continuous interval, a search and matching space between the nozzle's injection point and the printed dot matrix is ​​quickly constructed through spatial partitioning mapping. This effectively avoids the increased computational complexity caused by global traversal search, thereby reducing the complexity of large-size patterned planning and search matching. Based on this, the current nozzle-printing dot matching weight (i.e., priority weight) is determined by combining the actual injection point of each nozzle with the Y-axis positional deviation between each printed dot in the candidate space of that nozzle's printed dot, and the current nozzle injection state. This allows the matching weight to be adaptively adjusted, thereby improving the adaptability of the matching result to the actual injection characteristics. Furthermore, a maximum weight matching strategy is executed based on the priority weight. This invention can effectively improve the search and matching efficiency and printing effect in the thin-film encapsulation patterned printing process.

[0016] 2. This invention further identifies local conflict regions where nozzle resources compete, and uses a local maximum weight matching method to solve these local regions. By introducing local maximum weight correction, a one-to-one matching relationship between nozzles and printing points is determined. This method effectively establishes the local competition and conflict relationship between nozzle resources and printing points, guiding nozzle resources to be allocated to better printing points, thereby improving the overall matching effect and the final film quality while ensuring the feasibility of the printing process.

[0017] 3. This invention updates priority weights based on historical nozzle states and assigns matching weights to each candidate matching relationship, achieving adaptive weight optimization. This method constructs nozzle-printing point matching weights based on the spatial deviation between the nozzle ejection point and the printing point, combined with the current nozzle ejection state. A long short-term memory network is introduced to dynamically update the printhead ejection state, enabling the matching weights to adaptively adjust with historical printing results, thereby improving the adaptability of the matching results to actual ejection characteristics. Attached Figure Description

[0018] Figure 1 A flowchart illustrating a rapid search and matching method for nozzle-printing dots in inkjet patterning, provided in an embodiment of the present invention. Figure 2 A schematic diagram of the maximum printable continuous nozzle range provided in an embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the initial matching strategy and local matching conflict between nozzles and printing points provided in an embodiment of the present invention. Figure 4 This is a schematic diagram of local maximum weight matching provided in an embodiment of the present invention; Figure 5 A schematic diagram of spatial partitioning mapping provided for an embodiment of the present invention; Figure 6 This is a schematic diagram of nozzle state parameter prediction provided in an embodiment of the present invention; Figure 7 This is a schematic diagram of TFE patterning planning and printing provided in an embodiment of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0020] Example 1 A fast search and matching method for nozzle-printing dots in patterned inkjet printing, such as Figure 1 As shown, it includes: Based on the spatial adjacency of the actual spray landing points of each nozzle in the array, the distance between adjacent actual spray landing points is calculated. If the distance exceeds a preset breakage threshold, it is determined that there is a breakage at that distance position, thereby dividing the array of actual spray landing points of all nozzles into multiple continuous intervals. Among the multiple continuous intervals, the longest interval is selected as the printable maximum continuous nozzle interval during the printing process, which is used to participate in nozzle-printing point matching with the printing point. All print points in the preset print dot matrix, except for the first row, are mapped to the first row along a direction perpendicular to the Y direction to obtain a new print dot matrix. The new print dot matrix is ​​then continuously partitioned in the Y direction, with overlapping partitions. Each nozzle in the largest printable continuous nozzle interval is traversed. Taking the first print point in the Y direction of the new print dot matrix as the zero point coordinate in the Y direction, the ratio of the Y-coordinate value of the actual spray landing point corresponding to the nozzle to the minimum distance between adjacent points in the new print dot matrix is ​​calculated. Based on this ratio, the nearest neighbor print point to the landing point is found, thereby determining the print dot partition in which the nozzle can participate in printing. The Y-distance between the actual spray landing point corresponding to the nozzle and each print point in the print dot partition in the print dot partition in the print dot matrix is ​​calculated. The set of print points whose Y-distance is not greater than a preset threshold is used as the candidate space of print points for the nozzle. Based on the Y-axis positional deviation between the actual spray landing point of each nozzle and each print point in the candidate space of the nozzle print point, and the current spraying state of the nozzle, the priority weight for each print point in the candidate space of the nozzle to be printed by that nozzle is determined. The matching process between the nozzle and the print point is modeled as a constrained maximum weight matching optimization problem. The objective is to maximize the overall matching weight so that nozzles with smaller spatial deviations and better spraying states are selected first, thereby obtaining the overall optimal nozzle-print point matching relationship. The constraints include that each print point is printed by one nozzle and each nozzle prints at most one print point. The matching relationship between the nozzle and the print point is determined by solving the maximum weight matching optimization problem.

[0021] For patterned printing of large-size substrates, using a traversal search method based on small-size regions for nozzle and print point matching leads to a significant expansion of the search space as the printing area grows, greatly increasing the computational complexity of the matching planning process. This makes it difficult to meet the requirements of planning efficiency and production cycle time in large-size printing. Furthermore, existing technologies often construct candidate matching relationships based on the spatial relationship between nozzle ejection points and print points, failing to incorporate factors such as nozzle ejection status and nozzle ejection stability into the matching modeling process. In the matching solution stage, a sequential greedy matching strategy is typically used, causing nozzles matched earlier to be fixed in subsequent matching processes. This makes it difficult to make global or local corrections to the overall matching results, resulting in uneven nozzle resource allocation and difficulty in effectively resolving local matching conflicts, ultimately affecting the consistency and quality of the final printed film.

[0022] In this embodiment, the maximum printable continuous nozzle range is determined based on the state (i.e., the position) of all nozzle landing points. This range is used to constrain the matching range between nozzles and printing points, thereby reducing the complexity of large-size patterning planning and search matching. Under the constraint of the maximum continuous nozzle range, considering the large number of nozzles and printing points in the thin-film encapsulation patterning planning process, a spatial partitioning mapping method is adopted to index and map the nozzles and printing points, constructing a search matching space for nozzles and printing points (i.e., the printing point partition where each nozzle can participate in printing). A candidate matching relationship (i.e., a candidate space for printing points for each nozzle) is established between the nozzle orifices and the printing points. Based on the candidate matching relationship, nozzle state parameters are introduced. According to the Y-axis position deviation between the actual spray landing point of each nozzle and each printing point in the candidate space of that nozzle, as well as the current spraying state of the nozzle, the priority weight for each printing point in the candidate space of that nozzle to be printed by that nozzle is determined. Based on the candidate matching relationship and matching weight, a maximum weight matching model is constructed. Matching decisions are performed on the printing points based on the matching weight to determine the corresponding preferred nozzle for the printing point. It should be noted that the nozzle is driven by a high-precision motion platform to complete uniform scanning in the X direction, while the position of the printing point in the Y direction is mainly determined by the physical arrangement of the nozzles in the nozzle array. Since the motion platform has high positioning accuracy and repeatability in the X direction and the scanning speed is stable, the X-axis landing point error can usually be accurately compensated by the motion control system (ignition timing), and its impact on the nozzle-printing point matching relationship is small. Therefore, this embodiment only considers the Y direction.

[0023] Therefore, this embodiment introduces a fast search and spatial constraint mechanism in the nozzle-print point relationship construction stage of the printing process, and introduces priority weights and improved maximum weight matching in the matching decision stage, achieving efficient search and high-quality matching of nozzle-print point relationships in large-size patterned printing scenarios. Specifically, this method uses a spatial partitioning mapping approach to index and map nozzle ejection points and print points, quickly constructing candidate correspondences between nozzles and print points before printing. This effectively avoids the increased computational complexity caused by global traversal search, significantly reducing the complexity of nozzle-print point combination search and enabling rapid retrieval of nozzle-print point relationships. Furthermore, this method constructs nozzle-print point matching weights based on the spatial deviation between nozzle ejection points and print points, combined with the current nozzle ejection state. This allows the matching weights to adaptively adjust with historical printing results, thereby improving the adaptability of the matching results to actual ejection characteristics.

[0024] In practical implementation, the maximum printable continuous nozzle range during the printing process can be achieved in the following way: Based on the spatial adjacency relationship of the actual spray landing points corresponding to each nozzle in the array, the distance between adjacent spray landing points is calculated, such as... Figure 2 As shown in the figure, the upper part represents the nozzle orifices. The actual spray points of all the orifices constitute the spray point state space. Orifices with spray points are marked in black, and those that cannot spray are marked in white. The Y-axis distance between adjacent spray points is calculated and denoted as... The distance is then compared with a preset disconnection threshold. When the distance between adjacent spray landing points exceeds the disconnection threshold, a breakpoint is determined at that location, thus dividing the spray landing point array corresponding to the nozzle into multiple continuous intervals. Among these intervals, the interval containing the interval length is selected. The longest interval is used as the maximum continuous interval, and areas outside the maximum continuous interval are not included in the search matching and printing plan.

[0025] Constraining the search for matching regions for nozzles and print points based on the maximum continuous interval is problematic because the nozzle and print point data is too large, and the traversal method is highly complex and severely impacts efficiency. This embodiment proposes a partitioned mapping search rule.

[0026] In practical implementation, the candidate space for printing points of each nozzle can be achieved in the following way: determining the first... The spray point of each nozzle is mapped to the zone. And calculate the spray landing point and the zone. The distance between each printed point; when the distance is not greater than, for example hour( To establish a candidate matching relationship between the nozzle ejection point and the corresponding printing point (the minimum adjacent distance in the new printing dot matrix), the above process is repeated for all nozzle ejection points to construct a candidate matching space between all nozzles and printing points.

[0027] In practice, the aforementioned priority weights are determined by the spatial deviation between the nozzle ejection point and the printing point, as well as the nozzle ejection state, and are expressed as follows: ,in For nozzles The jetting point and the printing point The Y-axis positional deviation between them is used to reflect the spatial geometric position of the nozzle relative to the target printing point. The degree of matching, Indicates nozzle The jetting state parameters are used to reflect the jetting performance characteristics of the nozzle during the actual printing process. The output mapping function can be linear or nonlinear depending on the nozzle. This priority weight is the evaluation basis for subsequent nozzle-printing point matching.

[0028] By introducing It can characterize the degree of positional matching between the actual jet landing point and the target printing point, thus prioritizing the selection of jets with smaller spatial deviations for printing, thereby reducing printing landing point errors and improving patterning accuracy. The landing point deviation can be normalized and represented, allowing for priority selection of nozzles with landing points closer to the center of the target print point during the matching process, thereby improving printing position accuracy. Simultaneously, by introducing nozzle ejection state parameters... It characterizes the dynamic performance characteristics of the nozzle during continuous printing. It can be determined by combining the nozzle's historical jetting data, current jetting behavior characteristics, and prediction models built based on historical data. It can reflect the performance differences of the nozzle in terms of jetting volume stability, jetting repeatability, and jetting consistency, so that nozzles with good jetting status are selected first during the matching process, and print quality is reduced due to nozzle performance fluctuations.

[0029] In practical inkjet printing systems, the nozzles within the same type of printhead exhibit consistency in geometry, drive method, and working environment, and their jetting behavior shows strong homology in statistical characteristics and evolutionary trends. Therefore, as a preferred implementation method, by aggregating historical jetting data from multiple nozzles within the same type of printhead for joint modeling, a unified jetting state prediction model applicable to that type of nozzle can be obtained, thereby avoiding the training costs and data sparsity problems associated with building a separate model for each nozzle.

[0030] In practice, for each printing point The determined optimal matching nozzle can be expressed as: .

[0031] In practical implementation, the maximum weight matching model consists of the following: Based on priority weights, the matching process between the nozzle orifice and the printing point is modeled as a constrained maximum weight matching optimization problem. By maximizing the overall matching weight, the nozzle with smaller spatial deviation and better spray state is selected first, thereby obtaining the overall optimal nozzle-printing point matching relationship. By solving the maximum weight matching optimization problem, the final matching relationship between the nozzle orifice and the printing point is determined.

[0032] One preferred implementation method is to introduce a binary matching variable. , used to indicate the first The nozzle and the first The matching relationship between each print point, when the nozzle... With print point During matching, ,otherwise The objective function of the maximum weight matching optimization problem is expressed as: This involves selecting a set of matching combinations from all candidate matches to maximize the sum of the selected matching weights, thereby prioritizing nozzle-printing point matching pairs with smaller spatial deviations and better jetting conditions globally. This also satisfies the following constraint: each printing point must be printed by one nozzle. Each nozzle can print at most one print dot. By solving the maximum weight matching model under the aforementioned constraints, a one-to-one optimal match between the nozzle orifice and the printing point can be achieved.

[0033] The maximum weighted matching problem designed in the above manner can obtain the nozzle-printed point correspondence with the optimal overall matching weight while maintaining the one-to-one allocation relationship between nozzles and printed points. This achieves a globally optimal matching result that balances printing accuracy and nozzle state stability. However, in stark contrast, in thin-film encapsulation patterning applications, the number of nozzles and printed points increases significantly. If the existing global traversal solution method is used, the computational complexity will rise rapidly, making it difficult to meet the computational efficiency requirements of actual printing processes.

[0034] As a preferred embodiment, the method further includes: Identify the matching relationship between nozzles and printing points. If the same nozzle is selected by multiple printing points in all matching relationships, determine all nozzles that can participate in printing at each of the multiple printing points. Based on the priority weight of each printing point selecting the corresponding nozzle for printing, construct a corresponding local nozzle-printing point matching problem. The goal of this problem is to maximize the matching weight of the local area, so that nozzles with smaller spatial deviations and better spraying states are selected first, thereby obtaining the locally optimal nozzle-printing point matching relationship. Under the condition of satisfying the constraints, perform local maximum weight matching optimization to solve the problem and determine the final matching relationship between nozzles and printing points in the local area.

[0035] In other words, this preferred embodiment further identifies local conflict regions where nozzle resources compete, and uses a local maximum weight matching method to solve for these local regions to determine the one-to-one matching relationship between nozzles and printing points. This method effectively establishes local competition and conflict relationships between nozzle resources and printing points, guiding nozzle resources to be allocated to better printing points, thereby improving the overall matching effect and final film quality while ensuring the feasibility of the printing process.

[0036] This preferred approach proposes to first execute an initial matching strategy, and then perform local maximum weight matching for local matching conflicts formed between nozzles and printing points in their spatial neighborhoods, so as to significantly reduce the overall solution complexity while ensuring matching quality.

[0037] Specifically, the priority weight between completing the nozzle and the print point. After calculation, the initial matching is based on the perspective of the printed point, and this matching is performed for each printed point. Independently determine its optimal candidate nozzle, such as Figure 3 As shown. For the print point The nozzle with the highest matching weight is selected as its initial matching target: By performing initial matching decisions based on optimal weights, an initial set of matching relationships from print points to nozzles can be quickly obtained. This initial matching process does not involve solving global constraints, has low computational complexity, and is suitable for quickly constructing candidate relationships in large-size matching scenarios.

[0038] However, because the initial matching process described above does not explicitly introduce a unique nozzle constraint, multiple print points may simultaneously select the same nozzle during the initial matching phase, resulting in nozzle resource competition within a local area. This phenomenon manifests as the same nozzle being assigned to multiple print points, violating the one-to-one matching constraint between nozzles and print points.

[0039] Based on this, in this preferred embodiment, after the initial matching is completed, conflict detection is performed on the matching results to identify local matching conflict regions where nozzle resource competition exists. Specifically, the set of nozzles that meets the following conditions is defined as the local matching conflict space: .

[0040] If at least two different print points select the same nozzle during the initial matching phase, then the nozzle and its corresponding set of print points are determined to constitute a local matching conflict region.

[0041] For each identified local matching conflict region, this embodiment reconstructs the nozzle-print point matching subgraph only within that local area. The matching subgraph consists of the conflicting nozzles and the print points with which they have candidate matching relationships, and its size is much smaller than the global nozzle-print point matching graph, thereby significantly reducing the complexity of subsequent solutions.

[0042] In the local matching subgraph, unique constraints on nozzles and print points are introduced. While maintaining the original matching weight definitions, a local maximum weight matching algorithm is used to optimize and solve the matching subgraph to determine the final one-to-one matching relationship between nozzles and print points within the local region. Figure 4 As shown.

[0043] By using the above method, this embodiment decomposes the maximum weight matching problem, which originally needed to be solved globally, into a hierarchical solution process of "global fast initial matching + local conflict detection + local maximum weight optimization". The computationally complex matching optimization operation is only performed in local areas where there is resource competition. This greatly improves the matching solution efficiency while ensuring the global consistency and optimality of the matching results. It is especially suitable for thin film encapsulation patterned inkjet printing application scenarios with a large number of nozzles and printing points.

[0044] In summary, this optimization method introduces an initial matching strategy based on optimal weights and local maximum weight correction to obtain the conflict relationship between nozzles and printing points in the spatial neighborhood. The local maximum weight matching method is used to determine the final matching result for the conflict area, making the nozzle resource allocation more reasonable and improving the overall printing effect.

[0045] In practical implementation, the above-mentioned local matching conflict resolution is as follows: In the matching process between nozzles and print points, local matching conflicts arise when the same nozzle is simultaneously selected by multiple print points, creating a nozzle resource conflict structure. Based on this local matching conflict scenario, the nozzle-print point local matching conflict space is determined. Construct the corresponding local nozzle-print point matching subgraph.

[0046] A local subgraph is defined as follows: ;in, Represents the set of conflict-related print points; Represents a set of conflict nozzles; This represents the set of candidate matching edges between the print point and the nozzle.

[0047] The local nozzle-print point matching subgraph is solved by using a local maximum weight matching algorithm to solve the matching subgraph under the premise of satisfying the uniqueness constraints of nozzles and print points, thereby determining the final matching relationship between nozzles and print points in the local area. Indicates the print point The optimal nozzle.

[0048] As a preferred implementation, the disconnection threshold is determined based on the target film-forming distance, the actual jetting point Y-axis deviation, and the allowable film-forming continuity deviation, and is used to distinguish whether film formation is continuous between adjacent actual jetting points.

[0049] As a preferred embodiment, if the preset printing dot matrix is ​​a uniformly spaced printing dot matrix, then the continuous partitioning method is as follows: Number the printed dots on the new printed dot matrix along the Y direction, and denote them as follows: , n is the total number of print dots on the new print dot matrix, and the spacing between adjacent print dots in the new print dot matrix is ​​denoted as n. Based on the numbering order of the print dots, the new print dot matrix is ​​divided into multiple local print dot partitions, each partition being denoted as... Define it as ; The method for determining the print dot partitioning that a nozzle can participate in printing based on the ratio is as follows: The ratio corresponding to each nozzle is rounded down and expressed as follows: In the formula, This indicates the nozzle The corresponding Y-coordinate value of the actual spray landing point is used to determine the nozzle. The actual spray point is mapped to the partition. That is, the nozzle The print dots that can participate in printing are divided into zones. Only search within that zone for nozzles Matching print points.

[0050] In this implementation method, during actual printing, due to the combined effects of nozzle manufacturing errors, droplet trajectory disturbances, and substrate movement errors, the center of the droplet landing point formed by the nozzle ejection is usually difficult to completely coincide with the center of the target printing point, and there is usually a certain spatial deviation. Experimental verification shows that when the deviation distance between the droplet landing point center and the printing point center does not exceed, for example, 1.2 times the minimum printing point spacing, it will not have a significant adverse impact on the final film quality and pattern consistency, and can still be judged as effective printing.

[0051] Based on the above determination range, it can be determined that one nozzle may match a maximum of 2-3 print points, thus forming potential matching relationships with multiple print points. To reduce the complexity of global matching and constrain the range of candidate relationships between nozzles and print points, this method is used to divide the print points into multiple local spatial regions, each region denoted as . , ,like Figure 5 As shown, it is defined as:

[0052] Number the nozzles and the spray landing points; the nozzles are denoted as... The corresponding spray point's coordinates relative to the first printed point in the Y direction of the new printed dot matrix are: ,in Based on the minimum center-to-center distance of adjacent printed dots in the Y direction in the newly mapped printed dot matrix. By discretizing the relative coordinates of the jet landing point, a continuous space is mapped to a pre-divided local spatial region of the printing point. Specifically, when the nozzle... The corresponding jet landing point satisfies When, then determine the first The spray point of the first jet is mapped to the second jet. Each print point is partitioned, and only in the partition Find a matching print point in the middle.

[0053] In this preferred embodiment, considering jetting deviation, the actual printed point corresponding to the jetting landing point can only appear within a finite neighborhood. Combined with the candidate matching distance constraint:

[0054] It is known that when the jetting point deviates to its maximum, the number of printable points that can be matched will only expand to the range of adjacent printable points. Therefore, the search space is limited to include... Local spatial partitioning This can cover the entire set of print dots that may correspond to the jet landing point, and ensure that no valid matching relationships are missed. In a preferred implementation, the preset threshold is determined based on the minimum distance between adjacent print dots in the new print dot matrix and the spreading properties of the ink to be printed. The formula is selected as follows. This is just one implementation example.

[0055] By adopting a spatial mapping determination method, not only are complex distance calculations in continuous space avoided, but also a rapid determination of the partition from the nozzle landing point to the printing point is achieved, which effectively reduces the computational complexity of matching search. After completing the local spatial mapping between the nozzle landing point and the printing point, each printing point usually corresponds to multiple candidate nozzles, and the same nozzle may also form a potential matching relationship with multiple printing points.

[0056] However, relying solely on spatial mapping relationships is insufficient to determine the optimal matching relationship between nozzles and print points. Therefore, based on the aforementioned spatial mapping and candidate relationship constraints, this embodiment further introduces priority weights to evaluate the quality of nozzle-print point matching relationships, in order to quantitatively describe different nozzle-print point combinations and provide a unified evaluation basis for subsequent rapid search matching and conflict resolution.

[0057] Priority weight Established on the candidate relationship space between printhead orifices and printed dots, used to characterize printed dots. Select spray nozzle The priority of spraying. The priority weight comprehensively reflects the degree of spatial matching between the nozzle spray point and the print point, as well as the impact of the current spraying state of the nozzle on printing stability. It is determined by the spatial deviation between the nozzle spray point and the print point and the current spraying state of the printhead.

[0058] As a preferred embodiment, the current spray state of each nozzle can be determined by the following method: Obtain the jetting deviation sequence of the nozzle in the first t printing cycles, wherein the jetting deviation sequence includes the Y-direction position deviation between the actual jetting point of the nozzle and the target position, the jetting volume deviation and / or other temporal indicators that affect the jetting state; The jetting deviation sequence is input into a pre-trained Long Short-Term Memory (LSTM) network to predict the possible current jetting state of the nozzle in subsequent printing.

[0059] Considering that by jointly modeling historical jetting data from multiple nozzles of the same type, a unified jetting state prediction model applicable to this type of nozzle can be obtained, this preferred method, for any nozzle... Its state set is During the printing process, the first step During the next injection, construct the nozzle. The instantaneous state vector, including the spatial deviation between the nozzle impact point and the target position, the jet volume deviation, or other temporal indicators that may affect the jet state, can be expressed as: .in This indicates the spatial deviation of the nozzle's impact point in the X and Y directions relative to the theoretical target position. This indicates the volume deviation of the orifice during this injection.

[0060] For the nozzle At the current moment Previously, a length of 1 was selected. The historical state sequence constitutes This historical sequence is used to describe the trend of nozzle jetting behavior over time.

[0061] For example, based on historical data of all nozzles of the same type, a training sample set is constructed to train a Long Short-Term Memory (LSTM) network applicable to the prediction of this type of nozzle. The network uses the spatial deviation between the nozzle's impact point and the target position, and the jet volume deviation, as inputs, with subsequent jet state predictions as outputs. This example uses the LSTM network to model the historical state sequence of nozzles and predict nozzle state parameters, such as... Figure 6 As shown. For any nozzle The historical state sequence is input into the LSTM network time by time, and the network selectively remembers and forgets the historical information through a gating mechanism.

[0062] In the At time t, the hidden state update of LSTM is represented as ( Based on the final hidden state, the output jet state prediction result is determined. ,in, The output mapping function can be either a linear mapping or a nonlinear mapping.

[0063] This preferred embodiment updates the priority weights based on historical nozzle states and assigns matching weights to each candidate matching relationship, achieving adaptive weight optimization. This method constructs nozzle-printing point matching weights based on the spatial deviation between the nozzle ejection point and the printing point, combined with the current nozzle ejection state. A long short-term memory network is introduced to dynamically update the printhead ejection state, enabling the matching weights to adaptively adjust with historical printing results, thereby improving the adaptability of the matching results to actual ejection characteristics.

[0064] In summary, this embodiment proposes a search and matching method suitable for large-size patterned printing scenarios. First, the maximum printable continuous interval of the nozzles during the printing process is determined. Under the constraint of this continuous interval, a spatial partitioning mapping mechanism is introduced to construct the relationship between nozzles and printing points, thereby effectively reducing the size of the matching search space. In the matching solution stage, considering both the nozzle ejection state information and the matching weight relationship between nozzles and printing points, candidate nozzles are first assigned to each printing point. Based on this, local matching conflicts formed between nozzles and printing points in their spatial neighborhood are identified. For these local conflict areas, a local maximum weight correction mechanism is introduced. Under the premise of satisfying the uniqueness constraints of nozzles and printing points, the matching relationship is optimized and adjusted to determine the one-to-one final matching result between nozzles and printing points. This method significantly reduces the complexity of nozzle-printing point combination search, improves search and matching efficiency, and effectively improves the rationality of nozzle resource allocation, thereby enhancing the overall matching effect and the consistency and quality of the final film.

[0065] Example 2 A patterned inkjet printing system is provided, which is equipped with an inkjet planning module. The inkjet planning module uses the nozzle-printing point fast search and matching method as described in Embodiment 1 to realize the correspondence between the printing point and the nozzle in the Y-direction spatial position.

[0066] like Figure 7 As shown, the patterned planning and printing system includes at least: an information acquisition module, a printing planning module, and a printing module.

[0067] The information acquisition module collects the spatial distribution of the ejection points of each nozzle, the volume of the ejected ink droplets, and the nozzle angle; and collects the spatial coordinates of the dots to be printed on the substrate, the substrate tilt angle, and the target printing volume value corresponding to each printing point. The information acquisition module is used to collect nozzle status information and print point target information during the trial printing process, providing basic data support for subsequent matching planning and printing control.

[0068] Specifically, the information acquisition module first controls the printhead to perform a trial print operation based on the spatial arrangement of the printhead and its nozzles. It then collects the spatial position distribution of each nozzle's ejection point in the X and Y directions of the substrate coordinate system, thereby constructing a state-space distribution model of the nozzle ejection points for selection of the maximum continuous interval during the planning process. Simultaneously, it needs to collect the ejected droplet volume and printhead angle to plan the number of prints and trigger timing.

[0069] Specifically, the information acquisition module is used to acquire the spatial coordinate information of the dots to be printed on the substrate, including the X and Y coordinates of the printing dots in the substrate coordinate system, the substrate tilt angle, and the target printing volume or target ink volume requirement corresponding to each printing dot. The above printing dot information serves as the target constraint condition for nozzle-printing dot matching and jetting control.

[0070] Through the above methods, the information acquisition module achieves unified collection and organization of nozzle-side status information and print point-side target information, providing input data for the matching decision of the subsequent printing planning module.

[0071] The printing planning module is used to determine the optimal planning scheme based on the data provided by the information acquisition module, using the method as described in Example 1, and to complete the matching of nozzles and printing points; it also completes the division of printing intervals and the planning of the jetting sequence, thereby realizing the overall planning and control of thin film encapsulation patterned printing.

[0072] Specifically, the printing planning module first determines the maximum continuous printing interval along the substrate's movement direction based on the spatial distribution of the printhead nozzles and the substrate's movement direction. The maximum continuous printing interval represents the spatial range within which the printhead can complete printing without switching printing areas during a single continuous substrate scan. This limit is used to constrain the matching search space and reduce overall planning complexity. Printhead-printing point search and matching are performed within the maximum continuous interval. The search and matching related technical solutions are the same as in Embodiment 1 and will not be repeated here.

[0073] After completing the mapping of the print points and nozzles in the Y-axis spatial position, the print planning module further plans the jetting timing for the substrate's scanning motion along the X-axis. Specifically, using the planned baseline nozzle-print point mapping as a time reference, and combining the positional deviation of each nozzle's print landing point in the X-axis and the substrate's attitude change information, the module calculates the jetting trigger time for each nozzle, thereby forming a jetting timing planning result that matches the substrate's scanning motion.

[0074] After completing the matching and timing planning of all print points within a single maximum continuous print interval, the print planning module redetermines the maximum continuous print interval for the next area to be planned based on the new interval start print point and nozzle spatial relationship, and repeats the above planning process until a complete patterned print planning file covering the entire area to be sprayed on the substrate is generated.

[0075] The printing module includes a printhead control unit and a motion control unit. The printhead control unit comprises a printhead control board and a printhead drive board, while the motion control unit includes a base X-axis motion platform and a printhead Y-axis motion platform. It generates print data files and motion platform parameter data, which are then transmitted to the printhead control unit and motion control unit of the printhead module, respectively.

[0076] Specifically, as shown in Figure 2, the printhead motion platform moves along the Y-axis, and the substrate moves along the X-axis under the drive of the base. By controlling the timing of the printhead's spraying action, each nozzle completes spraying at the corresponding Y-axis position according to the planned timing during the X-axis scanning motion of the base. By repeating the above process, the spraying control of the entire printing process is completed.

[0077] The relevant technical solutions are the same as above, and will not be repeated here.

[0078] Example 3 This application also relates to a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.

[0079] Specifically, the memory may include high-speed random access memory, as well as non-volatile memory, such as hard disks, RAM, plug-in hard disks, smart media cards (SMC), secure digital (SD) cards, flash cards, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0080] The relevant technical solutions are the same as above, and will not be repeated here.

[0081] In summary, this invention proposes a fast search and matching method for nozzle-printed dots in patterned printing. By introducing a fast search and spatial constraint mechanism in the nozzle-printed dot relationship construction stage, and weight optimization in the matching decision stage, initial matching decisions are executed based on the optimal weights. Furthermore, a local maximum weight correction is introduced to address local matching conflicts between nozzles and printed dots in their spatial neighborhoods. This achieves efficient search and high-quality matching of nozzle-printed dot relationships in large-size patterned printing scenarios, obtaining near-globally optimal nozzle-printed dot matching results without relying on global traversal solutions, thereby improving printing efficiency and pattern consistency. In conclusion, this invention provides a comprehensive solution for large-size patterned search and matching, effectively addressing the aforementioned problems.

[0082] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for rapid search and matching of nozzles and printing dots for patterned inkjet printing, characterized in that, include: Based on the spatial adjacency of the actual spray landing points of each nozzle in the array, the distance between adjacent actual spray landing points is calculated. If the distance exceeds the preset disconnection threshold, it is determined that there is a breakpoint at the distance position, thereby dividing the array of actual spray landing points of all nozzles into multiple continuous intervals. The longest interval among the multiple consecutive intervals is selected as the maximum printable continuous nozzle interval of the printhead during the printing process. Map all print points in the preset print dot matrix except for the first row of print points to the first row along a direction perpendicular to the Y direction to obtain a new print dot matrix. The newly printed dot matrix is ​​continuously divided into sections along the Y direction, with overlap between the sections. Traverse each nozzle in the maximum printable continuous nozzle range, take the first print point in the Y direction in the new print dot matrix as the zero point coordinate in the Y direction, calculate the ratio of the actual spray landing point Y direction coordinate value of the nozzle to the minimum adjacent point distance in the new print dot matrix, find the nearest neighbor print point based on the ratio, and thus determine the print dot partition that the nozzle can participate in printing. Calculate the Y-direction distance between the actual jet landing point corresponding to the nozzle and each print point in the print point partition that can participate in printing. The set of print points whose Y-direction distance is not greater than a preset threshold is used as the print point candidate space of the nozzle. Based on the Y-direction positional deviation between the actual jet landing point of each nozzle and each printing point in the candidate space of the nozzle printing point, and the current jetting state of the nozzle, the priority weight of each printing point in the candidate space of the nozzle printing point for selecting the nozzle for printing is determined. The matching process between the nozzle orifice and the printing point is modeled as a constrained maximum weight matching optimization problem. The objective is to maximize the overall matching weight so that the nozzle with smaller spatial deviation and better spray state is selected first, thereby obtaining the overall optimal nozzle-printing point matching relationship. The constraints include that each printing point is printed by one nozzle and each nozzle prints at most one printing point. By solving the maximum weight matching optimization problem, the matching relationship between the nozzle orifice and the printing point is determined.

2. The rapid search and matching method for nozzle-printing dots as described in claim 1, characterized in that, The method also includes: Identify the matching relationship between nozzles and printing points. If the same nozzle is selected by multiple printing points in all matching relationships, determine all nozzles that can participate in printing at each of the multiple printing points. Based on the priority weight of each printing point selecting the corresponding nozzle for printing, construct a corresponding local nozzle-printing point matching problem. The goal of this problem is to maximize the matching weight of the local area, so that nozzles with smaller spatial deviations and better spraying states are selected first, thereby obtaining the locally optimal nozzle-printing point matching relationship. Under the condition of satisfying the constraints, perform local maximum weight matching optimization to solve the problem and determine the final matching relationship between nozzles and printing points in the local area.

3. The rapid search and matching method for nozzle-printing dots as described in claim 1, characterized in that, The disconnection threshold is determined based on the target film-forming spacing, the actual jetting point Y-axis deviation, and the allowable film-forming continuity deviation.

4. The rapid search and matching method for nozzle-printing dots as described in claim 1, characterized in that, The preset printing dot matrix is ​​a uniformly spaced printing dot matrix, and the continuous partitioning method is as follows: Number the printed dots on the new printed dot matrix along the Y direction, and denote them as follows: , n is the total number of print dots on the new print dot matrix, and the spacing between adjacent print dots in the new print dot matrix is ​​denoted as n. ; Based on the numbering order of the print dots, the new print dot matrix is ​​divided into multiple local print dot partitions, each partition being denoted as . Define it as ; The method for determining the print dot partitioning that a nozzle can participate in printing based on the ratio is as follows: The ratio corresponding to each nozzle is rounded down and expressed as follows: In the formula, This indicates the nozzle The corresponding Y-coordinate value of the actual spray landing point is used to determine the nozzle. The actual spray point is mapped to the partition. That is, the nozzle The print dots that can participate in printing are divided into zones. .

5. The rapid search and matching method for nozzle-printing dots as described in claim 1, characterized in that, The preset threshold is determined based on the minimum spacing between adjacent print dots in the new print dot matrix and the spreading performance of the ink to be printed.

6. The rapid search and matching method for nozzle-printing dots as described in claim 1, characterized in that, The current spray status of each nozzle is determined in the following way: Obtain the jetting deviation sequence of the nozzle in the first t printing cycles, wherein the jetting deviation sequence includes the Y-direction position deviation between the actual jetting point of the nozzle and the target position, the jetting volume deviation and / or other temporal indicators that affect the jetting state; The jetting deviation sequence is input into a pre-trained Long Short-Term Memory (LSTM) network to predict the possible current jetting state of the nozzle in subsequent printing.

7. The rapid search and matching method for nozzle-printing dots as described in claim 1, characterized in that, The objective function of the maximum weight matching optimization problem is: In the formula, Used to indicate nozzle With print point The matching relationship between them, when the nozzle With print point During matching, ,otherwise ; Indicates the print point Select spray nozzle Print priority matches weight; N Indicates the total number of print dots. M This indicates the total number of nozzles.

8. A patterned inkjet printing system, characterized in that, It is equipped with a printing planning module, which uses the nozzle-printing point fast search and matching method as described in any one of claims 1 to 7 to realize the correspondence between the printing point and the nozzle in the Y-direction spatial position.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein the computer program, when executed by a processor, controls the device on which the storage medium is located to perform the steps of the method as described in any one of claims 1 to 7.

Images