A data-driven inkjet printing droplet volume control method and system

Abstract

The present invention belongs to the field of printing display technology, and specifically relates to a data-driven inkjet printing ink droplet volume control method and system, including: using a nozzle to spray ink droplets under different waveform parameters, and obtaining the size of the printed ink droplet volume, constructing a mapping relationship between the printed ink droplet volume and the waveform parameters, and simultaneously constructing a fuzzy control system for the printed ink droplet volume; obtaining the waveform parameters corresponding to the target ink droplet volume according to the mapping relationship as reference waveform parameters; using the reference waveform parameters to perform a trial print, and if the deviation between the actual ink droplet volume and the target ink droplet volume exceeds a set threshold, using the fuzzy control system to adjust the waveform parameters until the deviation is less than the set threshold, and then outputting the adjusted waveform parameters for formal printing. The present invention can obtain suitable waveform parameters more quickly and accurately, and has a certain anti-disturbance capability, and is suitable for use in applications such as manufacturing high-resolution displays and electronic components using inkjet printing.

Description

Technical Field

[0001] The present invention belongs to the field of printing display technology, and more specifically, relates to a data-driven inkjet printing ink drop volume control method and system. Background Art

[0002] Inkjet printing is an emerging technology for manufacturing OLED displays. It precisely fills pixel pits on a substrate with ink droplets of a specific volume, enabling the printed production of displays. Compared to the commonly used vacuum thermal evaporation technology, inkjet printing offers advantages such as high material utilization, wide applicability, simple process, and low cost. It is ideally suited for manufacturing large-scale, high-precision OLED displays, and is currently the mainstream development direction for OLED display manufacturing.

[0003] The industrial application of inkjet printing technology in OLED display manufacturing still faces many challenges, one of which is how to accurately and quickly produce ink droplets of a specific volume. Failure to control the droplet size to the target value during inkjet printing can result in substandard ink volume within the pixel pits, leading to uneven printing and poor performance of the printed OLED functional layer. Therefore, precise and efficient control of droplet volume can significantly improve the efficiency and yield of inkjet printing in OLED display production.

[0004] In current industrial production, the most commonly used method for regulating ink droplet volume is the trial and error method, that is, manually changing the waveform parameters and measuring whether the ink droplet volume under the waveform meets the requirements. If not, continue to change the waveform parameters until the requirements are met. However, this method is time-consuming and labor-intensive, and wastes printing materials. If the adjustable range of the waveform parameters is extremely large, then the workload will be even more difficult to estimate when regulating the ink droplet volume to the target volume. Some scholars have established a prediction model between ink droplet volume and waveform parameters, which can save the tedious manual experiments. However, when the external environmental parameters change, the model is no longer reliable, and it is impossible to obtain ink droplets of the target volume in a timely and accurate manner. It does not have a certain anti-disturbance ability, and at this time, the trial and error method is still required for regulation. Summary of the Invention

[0005] In response to the defects of the existing technology and the need for improvement, the present invention provides a data-driven inkjet printing droplet volume control method and system, which aims to solve the technical problems that the existing droplet volume control methods have low accuracy and efficiency and are easily affected by environmental disturbances.

[0006] To achieve the above objectives, in a first aspect, the present invention provides a data-driven inkjet printing droplet volume control method, comprising:

[0007] Offline modeling stage:

[0008] Using a nozzle to eject ink droplets under different waveform parameters, and obtaining the volume of the printed ink droplets, constructing an inkjet printing ink droplet volume dataset, and constructing a mapping relationship between the printed ink droplet volume and the waveform parameters based on the dataset;

[0009] A fuzzy control system for the volume of inkjet droplets is constructed with the volume deviation of inkjet droplets as input and the change of waveform parameters as output.

[0010] Online volume control stage:

[0011] Obtaining waveform parameters corresponding to the target ink droplet volume according to the mapping relationship as reference waveform parameters;

[0012] A trial print is performed using the reference waveform parameters. If the deviation between the actual ink droplet volume and the target ink droplet volume is less than a set threshold, the reference waveform parameters are output for formal printing. Otherwise, the waveform parameters are adjusted using the fuzzy control system until the deviation is less than the set threshold, and the adjusted waveform parameters are then output for formal printing.

[0013] Furthermore, the inkjet printing ink drop volume dataset is constructed in the following manner:

[0014] In the waveform parameter adjustable range [(A d ,T d ),(A t ,T t )], randomly generate N groups of waveform parameters, use the ink drop observation system to measure the volume of the ink droplets under each group of waveform parameters, and record the inkjet printing ink droplet volume data set D; where A d and A t Respectively represent the lower and upper limits of the voltage amplitude, T d and T t Respectively represent the lower limit and upper limit of the voltage holding time;

[0015] The inkjet printing droplet volume dataset D is represented as {D=(V k ,A k ,T k )} k＝1,2...N , where V k A represents the volume of ink droplets printed by group k, k 、T k They respectively represent the voltage amplitude and voltage holding time corresponding to the volume of the ink droplets printed in the kth group.

[0016] Furthermore, the mapping relationship between the ink drop volume and the waveform parameters is constructed using a neural network method, specifically by constructing a network structure including m input layer neurons, n output layer neurons, and q hidden layer neurons, and using a Sigmoid function as an activation function;

[0017] The mapping relationship between the volume of the ink droplet and the waveform parameters is expressed as:

[0018]

[0019] where y j represents the jth output, x i represents the i-th input, v ih represents the network connection weight between the input layer neuron i and the hidden layer neuron h, w hj represents the network connection weight between hidden layer neuron h and output layer neuron j, θ j represents the threshold of the output layer neuron j, γ h represents the threshold of the hidden layer neuron h.

[0020] Furthermore, in the fuzzy control system, the input is the volume deviation of the ink droplet ΔV, and its value domain is ΔV∈[-ΔV t ,ΔV t ]; the output is the voltage amplitude change ΔA and the voltage holding time change ΔT, whose domains are ΔA∈[-a,a] and ΔT∈[-t,t] respectively; where ΔV t is the upper limit of volume deviation, a is the maximum step length of voltage amplitude change, and t is the maximum step length of voltage holding time change;

[0021] When the ΔV fuzzy subset is positive, the membership function

[0022] When the ΔV fuzzy subset is positive, the membership function

[0023] When the ΔV fuzzy subset is positive, the membership function

[0024] in,

[0025] When the ΔV fuzzy subset is negative large, negative medium, or negative small, its membership function is the membership function of the corresponding positive fuzzy subset, which is negative;

[0026] When the ΔT fuzzy subset is positive, the membership function

[0027] When the ΔT fuzzy subset is negative, the membership function

[0028] When the ΔT fuzzy subset is positive, the membership function

[0029] When the ΔT fuzzy subset is negative, the membership function

[0030] When the ΔA fuzzy subset is positive, the membership function

[0031] When the ΔA fuzzy subset is negative, the membership function

[0032] When the ΔA fuzzy subset is positive, the membership function

[0033] When the ΔA fuzzy subset is negative, the membership function

[0034] Where K is the adjustment coefficient.

[0035] Furthermore, the adjustment coefficient K is:

[0036]

[0037] Among them, V b is the current printing droplet volume, ΔV b is the current printing drop volume deviation, A a is the current voltage amplitude, T a is the current voltage holding time, A t is the upper limit of voltage amplitude, T t The upper limit of voltage holding time.

[0038] Furthermore, in the fuzzy control system, the fuzzy control rules are set as follows:

[0039] If the ΔV fuzzy subset is positive and large, then the ΔA fuzzy subset is negative and large, and the ΔT fuzzy subset is negative and large;

[0040] If the ΔV fuzzy subset is positive and neutral, then the ΔA fuzzy subset is negative and small, and the ΔT fuzzy subset is negative and large;

[0041] If the ΔV fuzzy subset is positively small, then the ΔA fuzzy subset is negatively small, and the ΔT fuzzy subset is negatively small;

[0042] If the ΔV fuzzy subset is negatively small, then the ΔA fuzzy subset is positively small, and the ΔT fuzzy subset is positively small;

[0043] If the ΔV fuzzy subset is negative medium, then the ΔA fuzzy subset is positive small, and the ΔT fuzzy subset is positive large;

[0044] If the ΔV fuzzy subset is negative and large, then the ΔA fuzzy subset is positive and large, and the ΔT fuzzy subset is positive and large. Further, the threshold λ is set to:

[0045] λ=V a *(c%-b%)

[0046] Where Va represents the target droplet volume, c% represents the maximum volume deviation percentage allowed in a single pixel pit, and b% represents the observed droplet volume deviation percentage.

[0047] In the second aspect, the present invention provides a printing method for printed display, which adopts the data-driven inkjet printing droplet volume control method described in the first aspect to determine the feasible value of the waveform parameter corresponding to the target droplet volume, and then uses the feasible value of the waveform parameter for formal printing.

[0048] In a third aspect, the present invention provides a data-driven inkjet printing droplet volume control system, the control system being configured to execute the data-driven inkjet printing droplet volume control method described in the first aspect, the control system comprising a nozzle module, a feedforward control module, a vision module, and a feedback control module;

[0049] The nozzle module is used to eject ink droplets;

[0050] The feedforward control module is used to obtain the waveform parameters corresponding to the target ink drop volume based on the mapping relationship between the ink drop volume and the waveform parameters, and complete the feedforward control of the ink drop volume;

[0051] The visual module is used to collect ink drop images to obtain the current printed ink drop volume;

[0052] The feedback control module is used to calculate the deviation between the current printed ink drop volume and the target ink drop volume. If the deviation exceeds a set threshold, the fuzzy control system is used to adjust the waveform parameters until the deviation is less than the set threshold, thereby obtaining a feasible value of the waveform parameters and completing feedback control of the printed ink drop volume.

[0053] In a fourth aspect, the present invention provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is executed by a processor, the device where the storage medium is located is controlled to execute a data-driven inkjet printing droplet volume control method as described in the first aspect, and / or a printing method for printed display as described in the second aspect.

[0054] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects:

[0055] (1) The feedforward-feedback control method provided by the present invention for the problem of ink droplet volume control can effectively improve the accuracy and efficiency of ink droplet volume control compared to existing methods, avoiding waste of manpower and material resources. At the same time, when the ink droplet volume changes due to some uncontrollable environmental factors, the method provided by the present invention can still use the constructed fuzzy control system to adjust the volume to the target volume by adjusting the waveform parameters, and has a certain anti-disturbance ability.

[0056] (2) The present invention sets an adjustment coefficient in the established fuzzy control system, so that the parameter adjustment amount can be reasonably adjusted according to the current state of the ink droplet volume, thereby adjusting the ink droplet volume to the target volume more quickly and avoiding excessive overshoot and material waste. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 This is one of the flow charts of the data-driven inkjet printing droplet volume control method provided by an embodiment of the present invention;

[0058] Figure 2 This is the second flow chart of the data-driven inkjet printing droplet volume control method provided by an embodiment of the present invention;

[0059] Figure 3 This is a schematic diagram of the usage status of the data-driven inkjet printing ink drop volume control system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0060] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is 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 for the purpose of explaining the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0061] In current industrial production, the most commonly used method for regulating ink droplet volume is the trial and error method, that is, manually changing the waveform parameters and measuring whether the ink droplet volume under the waveform meets the requirements. If not, continue to change the waveform parameters until the requirements are met. If the adjustable range of the waveform parameters is extremely large, then the workload will be even more difficult to estimate when regulating the ink droplet volume to the target volume. At present, some methods use a prediction model between the ink droplet volume and the waveform parameters to regulate the volume. Although this can save the tedious manual experiments, when the external environmental parameters change, the model is no longer reliable and cannot obtain the target volume of ink droplets in a timely and accurate manner. It does not have a certain anti-disturbance ability, and at this time, the trial and error method is still required for regulation. Therefore, a method for controlling ink droplet volume is proposed to address the above problems.

[0062] Example 1

[0063] See Figure 1 , combined with Figure 2 This embodiment provides a data-driven inkjet printing droplet volume control method, which includes an offline modeling stage and an online volume control stage. The offline modeling stage includes operations S1 and S2, and the online volume control stage includes operations S3 and S4.

[0064] Offline modeling stage:

[0065] Operation S1: Use the nozzle to spray ink droplets under different waveform parameters, obtain the volume of the printed ink droplets, construct an inkjet printing ink droplet volume data set, and construct a mapping relationship between the printed ink droplet volume and the waveform parameters based on the data set.

[0066] See Figure 3 , in the waveform parameter adjustable range [(A d ,T d ),(A t ,T t )], N groups of waveform parameters are randomly generated, and the ink droplet volume under each group of waveform parameters is measured using the ink droplet observation system, and the printed ink droplet volume data set is recorded.

[0067] The ink droplet volume dataset D includes waveform parameters and the corresponding ink droplet volume, that is, the dataset D can be expressed as {D=(V k ,A k ,T k )} k＝1,2...N , where V k A represents the volume of ink droplets printed by group k, k 、T k They respectively represent the voltage amplitude and voltage holding time corresponding to the volume of the ink droplets printed in the kth group.

[0068] Waveform parameter adjustable range [(A d ,T d ),(A t ,T t The upper and lower limits are determined by the hardware and existing control experience. The number of randomly generated waveform parameters, N, is determined by the adjustable range of the waveform parameters and should be no less than 20% of the total number of waveform parameter combinations within the adjustable range.

[0069] Using the inkjet printing droplet volume dataset D, a mapping relationship between the inkjet printing droplet volume and waveform parameters is constructed to perform feedforward control of the inkjet printing droplet volume.

[0070] The mapping relationship between the ink drop volume and the waveform parameters is constructed using a neural network method. The specific method is to construct a network structure including m input layer nodes, n output layer nodes, and q hidden layer nodes, where m = 1 and n is the number of categories included in the waveform parameters. In this embodiment, the waveform parameters include voltage amplitude and voltage holding time, that is, n = 2. The neural activation function uses the Sigmoid function. Use a small random number to initialize the network connection weight w between every two neurons and the threshold θ of each neuron.

[0071] 10%-20% of the samples in the inkjet droplet volume dataset D are randomly selected as the test set Dt, and the rest are used as the training set D1.

[0072] Use the training set D1 to train the neural network and update the connection weights in the neural network:

[0073]

[0074] Where w hj represents the network connection weight between the hidden layer neuron h and the output layer neuron j, η is the learning rate, E n For the network in a training example (x n ,y n ), β j is the input received by the j-th neuron in the output layer, For a training example (x n ,y n ) The output of the jth neuron in the neural network is determined as follows:

[0075]

[0076] Where θ j is the jth neuron in the output layer, For training examples (x n ,y n ) outputs the jth true value, b h is the output of the hth neuron in the hidden layer.

[0077] Will w hj After finishing, we can get:

[0078]

[0079] Use the trained neural network to predict the test set Dt and calculate the network output y j and actual Difference, get the prediction accuracy s of the model:

[0080]

[0081] Repeat the above process using training examples in the training set until the prediction model reaches the required accuracy. This completes the mapping relationship between the ink drop volume and the waveform parameters. The output obtained for any input sample x is:

[0082]

[0083] where y j Indicates the jth output quantity, that is, the jth waveform parameter category, x i represents the i-th input quantity, i.e. the volume of the printed ink drop; v ih represents the network connection weight between the input layer neuron i and the hidden layer neuron h, w hj represents the network connection weight between hidden layer neuron h and output layer neuron j, θ j represents the threshold of the output layer neuron j, γ h represents the threshold of the hidden layer neuron h.

[0084] In this way, the corresponding waveform parameters can be quickly obtained according to the target ink droplet volume, thereby completing feedforward control.

[0085] Operation S2 is to construct a fuzzy control system for the volume of ink droplets printed, using the deviation of the volume of ink droplets printed as input and the change in the waveform parameters as output.

[0086] Specifically, the construction method of the fuzzy control system is as follows: determine the input and output of the fuzzy control system, the value domain, and the membership function of each variable; set the fuzzy control rules of the system; perform fuzzy logic reasoning to obtain the fuzzy output; defuzzify the fuzzy output to obtain the precise value of the waveform parameter adjustment.

[0087] The input is the volume deviation of the ink droplet, ΔV, whose domain is ΔV∈[-ΔV t ,ΔV t ]; the output is the voltage amplitude change ΔA and the voltage holding time change ΔT, whose domains are ΔA∈[-a,a] and ΔT∈[-t,t] respectively; where ΔV t is the upper limit of volume deviation, a is the maximum step length of voltage amplitude change, t is the maximum step length of voltage holding time change, and ΔV t The values ​​of ,a, and t are determined by expert experience and the adjustable range of hardware equipment.

[0088] It is further necessary to select the corresponding membership function to convert ΔV∈[-ΔV t ,ΔV t] is mapped to fuzzy subsets, and six fuzzy subsets are defined here: negative large, negative medium, negative small, positive small, positive medium, and positive large; it is also necessary to select a membership function to map ΔA∈[-a,a],ΔT∈[-t,t] to fuzzy subsets, and four fuzzy subsets are defined here: negative large, negative small, positive small, and positive large.

[0089] When the ΔV fuzzy subset is positive, the membership function

[0090] When the ΔV fuzzy subset is positive, the membership function

[0091] When the ΔV fuzzy subset is positive, the membership function

[0092] Where, ΔV a The specific value is affected by factors such as inkjet printing equipment and ink type, and can be determined by expert experience. Generally, its value range is

[0093] The membership functions of the three negative fuzzy subsets of the ink drop volume deviation ΔV, namely, negative large, negative medium, and negative small, can be obtained by simply negating the membership functions of the corresponding positive fuzzy subsets.

[0094] When the ΔT fuzzy subset is positive, the membership function

[0095] When the ΔT fuzzy subset is negative, the membership function

[0096] When the ΔT fuzzy subset is positive, the membership function

[0097] When the ΔT fuzzy subset is negative, the membership function

[0098] The membership function of the voltage amplitude change ΔA is the same as the membership function of the voltage holding time change ΔT, except that t is replaced by a.

[0099] When the ΔA fuzzy subset is positive, the membership function

[0100] When the ΔA fuzzy subset is negative, the membership function

[0101] When the ΔA fuzzy subset is positive, the membership function

[0102] When the ΔA fuzzy subset is negative, the membership function

[0103] The membership relationship of ΔT and ΔA for the fuzzy subset is affected by the current printed ink drop volume, the current voltage amplitude, and the current voltage holding time. Therefore, an adjustment coefficient K needs to be introduced to adjust the membership relationship in real time to ensure that the membership relationship is more accurate, thereby ensuring the accuracy of subsequent adjustments.

[0104] According to the printed ink drop volume dataset D and the physical model of pressure wave propagation in piezoelectric inkjet, the adjustment coefficient K is obtained by fitting:

[0105]

[0106] Among them, V b is the current printing droplet volume, ΔV b is the current printing drop volume deviation, A a is the current voltage amplitude, T a is the current voltage holding time, A t is the upper limit of voltage amplitude, T t The upper limit of voltage holding time.

[0107] Establish appropriate fuzzy control rules:

[0108] According to adjustment experience, if the volume deviation ΔV is positive, the voltage amplitude change ΔA is negative, and the voltage holding time change ΔT is negative.

[0109] If the volume deviation ΔV is positive, the voltage amplitude change ΔA is small and negative, and the voltage holding time change ΔT is large and negative;

[0110] If the volume deviation ΔV is small and positive, the voltage amplitude change ΔA is small and negative, and the voltage holding time change ΔT is small and negative;

[0111] If the volume deviation ΔV is small and negative, the voltage amplitude change ΔA is small and positive, and the voltage holding time change ΔT is small and positive;

[0112] If the volume deviation ΔV is negative, the voltage amplitude change ΔA is positive and small, and the voltage holding time change ΔT is positive and large;

[0113] If the volume deviation ΔV is negative and large, the voltage amplitude change amount ΔA is positive and large, and the voltage holding time change amount ΔT is positive and large.

[0114] When the input quantity is determined, fuzzy logic reasoning can be performed according to the determined fuzzy control rules to obtain the fuzzy quantity. The specific methods are as follows:

[0115] If the volume deviation is ΔV1 at this time, the membership function calculation shows that the volume deviation belongs to the two fuzzy subsets of positive large and positive medium, and the membership is y1 and y2 respectively. Then, the fuzzy control rule can obtain the following results: when the volume deviation ΔV is positive and large, the voltage amplitude change ΔA is negative and large, and the voltage holding time change ΔT is negative and large; when the volume deviation ΔV is positive and medium, the voltage amplitude change ΔA is negative and small, and the voltage holding time change ΔT is negative and large. The next step is to use the Mamdani reasoning algorithm to calculate the fuzzy output, that is, to combine y1, y2 and the two results obtained by fuzzy logic reasoning. The final fuzzy output can be expressed as:

[0116]

[0117] in is the fuzzy output of the waveform parameter change, which is a vector of length 1*p, where p is the number of waveform parameter changes, B r The fuzzy input quantity representing the deviation of the ink drop volume, R r Obtain the input-output relationship for the p-th fuzzy logic reasoning.

[0118] Finally, the fuzzy output is defuzzified to obtain the precise value of the waveform parameter adjustment, thus completing the establishment of the fuzzy control system for the inkjet printing droplet volume.

[0119] Defuzzification can be performed using methods such as the area centroid method, the area equal division method, and the maximum membership average method. For example, the area centroid method is used for defuzzification, which is to find the center of the area enclosed by the fuzzy set membership function curve and the horizontal coordinate. The corresponding horizontal coordinate is the precise value of the waveform parameter adjustment.

[0120] Online volume control stage:

[0121] Operation S3: Obtain waveform parameters corresponding to the target ink droplet volume according to the mapping relationship as reference waveform parameters.

[0122] For example, the target ink drop volume set at this time is V a , V a Input into the established feedforward control, and the target ink drop volume V can be calculated according to the feedforward control model. a The corresponding waveform parameters (A a ,T a ), and input this waveform parameter as a reference waveform parameter into the inkjet printing device, and the nozzle ejects ink droplets to complete the feedforward control.

[0123] Operation S4 is to perform a trial print using the reference waveform parameters. If the deviation between the actual ink droplet volume and the target ink droplet volume is less than the set threshold, the reference waveform parameters are output for formal printing. Otherwise, the waveform parameters are adjusted using the fuzzy control system until the deviation is less than the set threshold, and the adjusted waveform parameters are output for formal printing.

[0124] For example, after feedforward control, the ink droplet volume V under the current waveform parameters is obtained by using a measurement method such as an ink droplet observation system. b , calculate its difference with the target ink drop volume V a If the deviation ΔV1 is less than the set threshold λ, the waveform parameters obtained by feedforward adjustment can be directly used for formal printing. If the value is greater than the set threshold λ, feedback adjustment is required.

[0125] The threshold λ is related to the allowable error of ink droplets and observation error, and can be set to

[0126] λ=V a *(c%-b%)

[0127] Where V a represents the target droplet volume, c% represents the maximum volume deviation percentage allowed in a single pixel pit, and b% represents the observed droplet volume deviation percentage.

[0128] The waveform parameters are adjusted by the feedback control system, and the waveform parameters are adjusted. The ink is ejected again using the adjusted waveform parameters. The volume is measured using the ink drop observation system. The difference ΔV2 from the target ink drop volume is calculated again and compared with the set threshold λ. If it is still greater than the threshold λ, the above steps are repeated until ΔV n <λ, record the waveform parameters (A n ,T n ). (A n ,T n ) is the feasible value of the waveform parameter corresponding to the target ink droplet volume, completing the feedback control of the ink droplet volume.

[0129] Example 2

[0130] A printing method for printed display uses a data-driven inkjet printing droplet volume control method as described in Example 1 to determine feasible values ​​of waveform parameters corresponding to target droplet volumes, and then uses the feasible values ​​of waveform parameters for formal printing.

[0131] Example 3

[0132] A data-driven inkjet printing drop volume control system, see Figure 3 , which includes a nozzle module, a feedforward control module, a vision module, and a feedback control module.

[0133] The feedforward control module 2 receives the input target ink droplet volume, obtains the corresponding waveform parameters using the established mapping relationship between the ink droplet volume and the waveform parameters, and transmits the waveform parameters to the host computer 6 .

[0134] The printhead module includes a printhead 31, a printhead driver card 32, an air pressure pump 33, and a print control card 34. The print control card 34 receives print data from the host computer 6 and generates corresponding waveform signals and negative pressure signals. The printhead driver card 32 receives the waveform signal transmitted from the print control card 34 and applies it to the printhead 31. The air pressure pump 33 receives the negative pressure signal from the print control card 23, adjusts the negative pressure, and ultimately forms ink droplets for ejection from the printhead 21.

[0135] The visual module includes a light source 41, an ink droplet observation camera 42, and an ink droplet observation system 43, wherein the lenses of the light source 41 and the ink droplet observation camera 42 are coaxially installed on both sides for collecting ink droplet images generated by the nozzle 21. The ink droplet observation system 42 receives the collected ink droplet images and obtains the current ink droplet volume after image processing.

[0136] The feedback control module includes a volume deviation calculation module 51 and a droplet volume fuzzy control system 52. These modules are used to calculate the deviation between the current droplet volume measured by the vision module and the target droplet volume. The volume deviation calculation module 51 receives the current droplet volume data and calculates the deviation between the current droplet volume and the target droplet volume. The droplet volume fuzzy control system 52 receives the droplet volume deviation and the current waveform parameters from the host computer. If the droplet deviation exceeds the set threshold λ, feedback adjustment is performed to obtain the waveform parameter adjustment amount, which is input to the host computer 6. The adjusted waveform signal is then input to the printing control card 34, thereby completing the droplet volume adjustment.

[0137] Example 4

[0138] A computer-readable storage medium, the computer-readable storage medium including a stored computer program, wherein when the computer program is executed by a processor, the device containing the storage medium is controlled to execute a data-driven inkjet printing droplet volume control method as described in Example 1, and / or a printing method for printed display as described in Example 2.

[0139] It will be easily understood by those skilled in the art 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 in the scope of protection of the present invention.

Claims

1. A data-driven inkjet printing droplet volume control method, characterized in that: include: Offline modeling stage: Using a nozzle to eject ink droplets under different waveform parameters, and obtaining the volume of the printed ink droplets, constructing an inkjet printing ink droplet volume dataset, and constructing a mapping relationship between the printed ink droplet volume and the waveform parameters based on the dataset; A fuzzy control system for the volume of inkjet droplets is constructed with the volume deviation of inkjet droplets as input and the change of waveform parameters as output. Online volume control stage: Obtaining waveform parameters corresponding to the target ink droplet volume according to the mapping relationship as reference waveform parameters; Performing a trial print using the reference waveform parameters; if a deviation between an actual ink droplet volume and a target ink droplet volume is less than a set threshold, outputting the reference waveform parameters for formal printing; otherwise, adjusting the waveform parameters using the fuzzy control system until the deviation is less than the set threshold, and then outputting the adjusted waveform parameters for formal printing; The inkjet printing droplet volume dataset is constructed in the following way: In the waveform parameter adjustable range [(A d ,T d ),(A t ,T t )], randomly generate N groups of waveform parameters, use the ink drop observation system to measure the volume of the ink droplets under each group of waveform parameters, and record the inkjet printing ink droplet volume data set D; where A d and A t Respectively represent the lower and upper limits of the voltage amplitude, T d and T t Respectively represent the lower limit and upper limit of the voltage holding time; The inkjet printing droplet volume dataset D is represented as {D=(V k ,A k ,T k )} k＝1,2...N , where V k A represents the volume of ink droplets printed by group k, k 、T k They respectively represent the voltage amplitude and voltage holding time corresponding to the volume of the ink droplets printed in the kth group.

2. The data-driven inkjet printing droplet volume control method according to claim 1, characterized in that: The mapping relationship between the ink drop volume and the waveform parameters is constructed using a neural network method, specifically by constructing a network structure including m input layer neurons, n output layer neurons, and q hidden layer neurons, and using a Sigmoid function as an activation function; The mapping relationship between the volume of the ink droplet and the waveform parameters is expressed as: where y j represents the jth output, x i represents the i-th input, v ih represents the network connection weight between the input layer neuron i and the hidden layer neuron h, w hj represents the network connection weight between hidden layer neuron h and output layer neuron j, θ j represents the threshold of the output layer neuron j, γ h represents the threshold of the hidden layer neuron h.

3. The data-driven inkjet printing droplet volume control method according to claim 1, characterized in that: In the fuzzy control system, the input is the volume deviation of the ink droplet ΔV, and its domain is ΔV∈[-ΔV t ,ΔV t ]; the output is the voltage amplitude change ΔA and the voltage holding time change ΔT, whose domains are ΔA∈[-a,a] and ΔT∈[-t,t] respectively; where ΔV t is the upper limit of volume deviation, a is the maximum step length of voltage amplitude change, and t is the maximum step length of voltage holding time change; When the ΔV fuzzy subset is positive, the membership function When the ΔV fuzzy subset is positive, the membership function When the ΔV fuzzy subset is positive, the membership function in, When the ΔV fuzzy subset is negative large, negative medium, or negative small, its membership function is the membership function of the corresponding positive fuzzy subset, which is negative; When the ΔT fuzzy subset is positive, the membership function When the ΔT fuzzy subset is negative, the membership function When the ΔT fuzzy subset is positive, the membership function When the ΔT fuzzy subset is negative, the membership function When the ΔA fuzzy subset is positive, the membership function When the ΔA fuzzy subset is negative, the membership function When the ΔA fuzzy subset is positive, the membership function When the ΔA fuzzy subset is negative, the membership function Where K is the adjustment coefficient.

4. The data-driven inkjet printing ink drop volume control method according to claim 3, characterized in that: The adjustment coefficient K is: Among them, V b is the current printing droplet volume, ΔV b is the current printing drop volume deviation, A a is the current voltage amplitude, T a is the current voltage holding time, A t is the upper limit of voltage amplitude, T t The upper limit of voltage holding time.

5. The data-driven inkjet printing ink drop volume control method according to claim 3 or 4, characterized in that: In the fuzzy control system, the fuzzy control rules are set as: If the ΔV fuzzy subset is positive and large, then the ΔA fuzzy subset is negative and large, and the ΔT fuzzy subset is negative and large; If the ΔV fuzzy subset is positive and neutral, then the ΔA fuzzy subset is negative and small, and the ΔT fuzzy subset is negative and large; If the ΔV fuzzy subset is positively small, then the ΔA fuzzy subset is negatively small, and the ΔT fuzzy subset is negatively small; If the ΔV fuzzy subset is negatively small, then the ΔA fuzzy subset is positively small, and the ΔT fuzzy subset is positively small; If the ΔV fuzzy subset is negative medium, then the ΔA fuzzy subset is positive small, and the ΔT fuzzy subset is positive large; If the ΔV fuzzy subset is negative and large, then the ΔA fuzzy subset is positive and large, and the ΔT fuzzy subset is positive and large.

6. The data-driven inkjet printing ink drop volume control method according to claim 1, characterized in that: The threshold λ is set to: λ=V a *(c%-b%) Where V a represents the target droplet volume, c% represents the maximum volume deviation percentage allowed in a single pixel pit, and b% represents the observed droplet volume deviation percentage.

7. A printing method for printed display, characterized in that: A data-driven inkjet printing droplet volume control method as described in any one of claims 1 to 6 is used to determine feasible values ​​of waveform parameters corresponding to target droplet volumes, and then the feasible values ​​of waveform parameters are used for formal printing.

8. A data-driven inkjet printing ink drop volume control system, characterized in that: The control system is used to execute the data-driven inkjet printing ink drop volume control method according to any one of claims 1 to 6, and the control system includes a nozzle module, a feedforward control module, a vision module, and a feedback control module; The nozzle module is used to eject ink droplets; The feedforward control module is used to obtain the waveform parameters corresponding to the target ink drop volume based on the mapping relationship between the ink drop volume and the waveform parameters, and complete the feedforward control of the ink drop volume; The visual module is used to collect ink drop images to obtain the current printed ink drop volume; The feedback control module is used to calculate the deviation between the current printed ink drop volume and the target ink drop volume. If the deviation exceeds a set threshold, the fuzzy control system is used to adjust the waveform parameters until the deviation is less than the set threshold, thereby obtaining a feasible value of the waveform parameters and completing feedback control of the printed ink drop volume.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a stored computer program, wherein, when the computer program is executed by a processor, the device where the storage medium is located is controlled to execute a data-driven inkjet printing droplet volume control method as described in any one of claims 1 to 6, and / or a printing method for printed display as described in claim 7.

Images