Ink-jet printing UV curing film morphology regulation and control method and system
Through inkjet printing UV curing technology, the agent model and pulse curing mode are used to solve the problems of insufficient film uniformity and curing quality in traditional film forming technology, and efficient and uniform film forming is achieved.
Patent Information
- Application Number
- CN202510054557.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-14
- Publication Date
- 2025-05-30
AI Technical Summary
Traditional film forming technology has shortcomings in uniformity and curing quality, resulting in uneven film thickness, many surface defects and low production efficiency.
The morphology regulation method of UV cured films is adopted for inkjet printing. By constructing a training sample set, training agent model, finding optimization, and using pulse curing mode and temperature control, uniform forming of the film is achieved.
The uniform molding of the film is achieved, with high-quality surface morphology and thickness consistency, and the problems of uneven film thickness, insufficient curing rate and surface defects are solved.
Smart Images
Figure CN120056315A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field related to thin film forming, and more specifically, relates to a method and system for regulating the morphology of an inkjet-printed UV-cured thin film. Background Art
[0002] In the context of the rapid development of today's technology, thin film technology has become an indispensable part of many high-tech fields. Due to its unique physical, chemical, and mechanical properties, thin films play a crucial role in many fields such as microelectronics, optical devices, solar cells, and biosensors. However, traditional thin film forming technologies, although meeting the requirements of industrial production to a certain extent, still have some insurmountable defects and deficiencies, especially in terms of the uniformity and curing quality of thin films.
[0003] Traditional thin film forming methods mainly include spin coating, dip coating, spraying, and other techniques. When preparing thin films using these methods, they often rely on manual operations or simple mechanical devices, making it difficult to precisely control the thickness and uniformity of the thin films. For example, during spin coating, factors such as the viscosity of the solution, the rotation speed, and the time will affect the final quality of the thin film. If the operation is improper, it is easy to cause problems such as uneven film thickness, edge curling, or central depression. Similar problems also exist in the dip coating method, where the distribution of the solution on the substrate is uneven, and holes or thickness differences are likely to occur after curing. Although the spraying technique can achieve the preparation of large-area thin films, its atomization effect and spraying uniformity are greatly affected by the equipment performance and operation skills.
[0004] In addition to the uniformity problem, traditional thin film forming methods also face many challenges during the curing process. Many materials need to be thermally cured or photo-cured under specific conditions, which is often time-consuming and energy-intensive. In some cases, long-term thermal curing may also lead to the degradation of material properties, affecting the final performance of the thin film. Although photo-curing technology has the advantages of fast curing speed and low energy consumption, its curing depth and uniformity are limited by the light source intensity and wavelength, making it difficult to achieve uniform curing of thick films or substrates with complex shapes.
[0005] In high-precision application fields such as microelectronics and optical devices, the uniformity and quality of thin films are directly related to the performance and reliability of the devices. For example, in integrated circuit manufacturing, the uniformity of thin films directly affects the conductive performance and signal transmission quality of the circuit. In optical devices, the refractive index and thickness uniformity of thin films determine the imaging quality and light energy distribution of the optical system. Therefore, traditional thin film forming technologies can no longer meet the stringent requirements for thin film quality in these fields.
[0006] In addition, with the progress of technology and the development of the market, the requirements for the performance of thin films are constantly increasing. The continuous emergence of new thin film materials poses higher challenges to thin film forming technologies. These materials often have special chemical compositions and molecular structures, and require more refined and professional curing processes to ensure the full play of their performance. Traditional thin film forming methods often seem inadequate when dealing with these new materials.
[0007] Although existing technologies can meet the requirements of thin film forming to a certain extent, there are still problems such as uneven thin film thickness, many surface defects, and low production efficiency in practical applications. Therefore, it is of great significance to develop a control method that can form thin films efficiently and uniformly. Summary of the Invention
[0008] In view of the above defects or improvement requirements of the existing technology, the present invention provides an inkjet printing UV curing thin film morphology regulation method and system, aiming to propose a control method that can form thin films efficiently and uniformly to solve the problems of uneven thin film thickness, insufficient curing rate, and surface defects existing in the existing technology.
[0009] To achieve the above object, according to one aspect of the present invention, an inkjet printing UV curing thin film morphology regulation method is provided, including:
[0010] S1. Construct a training sample set, where each training sample includes a value combination of multiple process parameters as the input of the model and index values of multiple indicators reflecting the curing effect as the label; the multiple process parameters are parameters determined by screening and meeting the preset requirements for affecting the performance of the cured thin film;
[0011] S2. Use the training sample set to train multiple surrogate models between the value combinations of process parameters and the index values; perform weighted averaging on the output ends of the surrogate models whose prediction accuracy meets the preset requirements to form a performance evaluation surrogate model;
[0012] S3. Use a heuristic sorting algorithm to perform optimization respectively with the target that the index values of each index output by the performance evaluation surrogate model meet the preset requirements, and determine the Pareto front solution sets corresponding to the satisfaction of each index with the preset requirements; intersect the Pareto front solution sets to obtain a process parameter compromise solution set that meets the preset requirements of all indexes;
[0013] S4. According to actual needs, determine the value combination of the multiple process parameters from the process parameter compromise solution set for curing regulation.
[0014] Further, the method for constructing the training sample set is:
[0015] The orthogonal experimental design method is adopted to obtain sample data through curing experiments, and the process parameters are screened based on range analysis and variance analysis. The process parameters that meet the preset requirements for the performance of the cured film are retained as the input of the surrogate model.
[0016] Among them, the process parameters involved in the screening include: UV power, radiation intensity, light uniformity, light exposure time, light source moving speed, and surface energy.
[0017] Furthermore, the multiple indicators include curing rate, curing uniformity, and film morphology.
[0018] Furthermore, the multiple surrogate models include RSM model, RBF model, Kriging model, ANN model, SVM model, and / or RF model.
[0019] Furthermore, the screening method for the surrogate model with prediction accuracy meeting the preset requirements is as follows:
[0020] Calculate the R 2 and RMSE values between the predicted values and the true values of each surrogate model, and retain the single surrogate model with R 2 value greater than 0.9 and RMSE value less than 0.1;
[0021] Then, the retained surrogate models are combined using the weighted average method, and the expression of the performance evaluation surrogate model obtained is where w i is the weight parameter, and its value is determined according to the R 2 weighted value corresponding to the i-th retained surrogate model; N represents the number of retained surrogate models; is the output of the i-th retained surrogate model; represents the output of the performance evaluation surrogate model.
[0022] Furthermore, the S2 also includes: taking the process parameter value space corresponding to the largest output difference of the multiple surrogate models as the most uncertain region, and constructing a new training sample set in this most uncertain region to continue training the multiple surrogate models.
[0023] Furthermore, in the S4, the pulse curing mode is adopted, and the UV light intensity is increased step by step and UV light exposure is performed intermittently to achieve curing.
[0024] Furthermore, it also includes:
[0025] S5. Real-time obtain sample data including process parameter value combinations and their corresponding index values, apply the utility criterion and the addition point criterion to select sample data for optimizing and improving the surrogate model with prediction accuracy meeting the preset requirements, and repeat the execution of S3.
[0026] According to another aspect of the present invention, an inkjet printing UV-curing film topography regulation system is provided for implementing an inkjet printing UV-curing film topography regulation method as described above, including a UV light source, an infrared interferometer, a surface topography analyzer, a temperature sensor, a differential pressure sensor, a UV energy meter, an ultrasonic sensor, a light source control module, and a temperature control module;
[0027] Among them, the UV light source is used to provide the light required for curing; the infrared interferometer and the surface topography analyzer are used to effectively measure and record the curing rate, uniformity, thickness, and topography of the cured film offline to obtain the index values of various indicators; the temperature sensor, the differential pressure sensor, the UV energy meter, and the ultrasonic sensor are used to record the process and environmental parameter data of the curing process in real time; the light source control module is used to adjust the output intensity of the UV light source according to actual needs; the temperature control module is used to monitor the temperature of the curing area in real time through the temperature sensor and perform temperature control.
[0028] Further, the UV light source is an arrayed ultraviolet LED light source; the temperature control module performs temperature control in a multi-point temperature control manner.
[0029] Generally speaking, compared with the prior art through the above technical solutions conceived by the present invention, the technical solutions provided by the present invention mainly have the following beneficial effects:
[0030] 1. The present invention proposes an inkjet printing UV-curing film topography regulation method, which uses various indicators reflecting the curing effect as the output of the model and the combination of process parameter values as the input to train the surrogate model. Among them, the process parameters used as the model input are the parameters determined through screening and having a relatively high impact on film curing, which improves the regulation efficiency while reducing the input dimension of the model. In addition, multiple surrogate models are trained, and the surrogate model with high prediction accuracy is selected and weighted averaged to further ensure the prediction accuracy. After obtaining the comprehensive surrogate model, a multi-objective function is established based on multiple indicators, and an intelligent optimization algorithm is used to determine the solution sets of process parameters corresponding to each objective. Finally, each solution set is solved to determine the final solution set of process parameters that meet the requirements of each indicator, ensuring that the film can meet the requirements of each indicator during the curing process. Therefore, the method of the present invention can achieve uniform forming and has a high-quality surface topography and thickness consistency, and this method has wide applications in the fields of materials science, electronic manufacturing, and optical device manufacturing.
[0031] 2. Further, in order to avoid problems such as film shrinkage and uneven heat absorption caused by long-term curing at one time and improve the uniformity of the film, the present invention adopts a pulsed curing mode. By gradually increasing the UV light intensity or gradually extending the irradiation time in steps, the film is cured from the surface to the inside, reducing internal stress; and combined with intermittent UV light irradiation to control the curing process of the film. This method can effectively reduce heat accumulation, avoid overheating and deformation of the film, and help improve the density of the film. Combined with temperature control, it can effectively reduce the influence of heat on the film and prevent thermal deformation of the film. BRIEF DESCRIPTION OF THE DRAWINGS
[0032] Figure 1 is a flowchart of a method for regulating the morphology of an inkjet-printed UV-cured film provided by an embodiment of the present invention;
[0033] Figure 2 is a schematic diagram of the correlation coefficient matrix of the remaining parameters and the film morphology index after screening by range analysis provided by an embodiment of the present invention;
[0034] Figure 3 is a schematic diagram of the orthogonal experimental design process provided by an embodiment of the present invention;
[0035] Figure 4 is a schematic diagram of the process for screening, optimizing, and combining multiple surrogate models provided by an embodiment of the present invention;
[0036] Figure 5 is an overall flowchart of a method for regulating inkjet-printed UV-cured film provided by an embodiment of the present invention;
[0037] Figure 6 is a schematic diagram of the configuration of an inkjet-printed UV-cured film morphology regulation system provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS
[0038] In order to make the objectives, technical solutions, and advantages of the present invention clearer, the present 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 only used to explain the present invention and are not used to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.
[0039] Example 1
[0040] An inkjet-printed UV-cured film morphology regulation method, as Figure 1 shown, includes:
[0041] S1. Construct a training sample set, where each training sample includes a combination of values of multiple process parameters as the input of the model and the index values of multiple indicators reflecting the curing effect as the labels; the multiple process parameters are the parameters determined by screening and meeting the preset requirements for the influence on the performance of the cured film.
[0042] S2. Use the training sample set to train multiple surrogate models between the combination of process parameter values and the index values; perform weighted averaging on the output ends of the surrogate models whose prediction accuracy meets the preset requirements to form a performance evaluation surrogate model.
[0043] S3. Use a heuristic sorting algorithm to perform optimization respectively with the goal that the index values of each index output by the performance evaluation surrogate model meet the preset requirements, and determine the Pareto front solution sets corresponding to each index meeting the preset requirements; intersect the Pareto front solution sets to obtain a trade-off solution set of process parameters that meets the preset requirements of all indexes.
[0044] S4. According to actual needs, determine the combination of values of the multiple process parameters from the trade-off solution set of process parameters for curing regulation.
[0045] In step S3 of the method of this embodiment, the objective functions of the evaluation surrogate models for each index are respectively established, and through the heuristic sorting algorithm and separately optimizing the multi-objective functions, the respective Pareto front solution sets are obtained. Regarding the intersection of the Pareto front solution sets in S3, the process parameters corresponding to the obtained multiple groups of Pareto front solution sets are summarized to form a process parameter set; it is judged whether the process parameter set is within the process parameter interval of the qualified curing degree, and the process parameters not within the interval are excluded; the unexcluded process parameters within the interval intersection are continuously judged, and it is judged whether the remaining process parameters are within the process parameter interval of curing uniformity, and the process parameters not within the range are continuously excluded; the unexcluded process parameters within the interval intersection are continuously judged, and it is judged whether the remaining process parameters are within the process parameter interval of the qualified film morphology, and the process parameters not within the range are continuously excluded to obtain the final recommended trade-off solution set of process parameters.
[0046] In addition, film curing does not involve droplet evaporation. The essential principle is that light acts as a catalyst, causing the liquid to undergo a polymerization reaction, which is a process of converting monomers into polymers.
[0047] As a preferred implementation manner, the above-mentioned manner of constructing the training sample set is as follows:
[0048] Adopt the orthogonal experimental design method, obtain sample data through curing experiments, and screen the process parameters based on range analysis and variance analysis, exclude the process parameters with less influence on the performance of the cured film, and retain the process parameters that meet the preset requirements for the influence on the performance of the cured film as the input of the surrogate model.
[0049] Among them, the process parameters involved in screening include: UV power, radiation intensity, light uniformity, light exposure time, light source moving speed, and surface energy.
[0050] The experimental method of the orthogonal table is used for parameter selection and data collection. In parameter screening, the importance of process parameters is analyzed by using range analysis, and the parameter with an R j with a too large gap from the mean value is removed. Then, the analysis of variance can effectively analyze the influence of the remaining multiple factors and their interactions on the experimental results. By using the F-test method and performing a one-sided test, it can be judged that this factor has no influence on the experimental results.
[0051] Specifically, the experimental method of selecting the orthogonal table in the form of L m (n k ) is used for screening, where: L is the orthogonal table symbol; m is the number of experimental groups; n is the number of levels; k is the number of factors.
[0052] Then, the factor with an R j with a too large gap from the mean value is removed by using range analysis, and its calculation formula is represents the average value of the indicators corresponding to each level of factor j, x i represents the response index value under each test, T represents the number of experiments under each level, R j represents the range of the indicators corresponding to factor j. If the range of a certain factor is very small, it means that this factor has little influence on the experimental results and can be considered to be ignored, thus simplifying the experimental design, represents the mean value of the indicators corresponding to each level of factor j.
[0053] Then, the analysis of variance can effectively analyze the influence of the remaining multiple factors and their interactions on the experimental results. By using the F-test method and performing a one-sided test, its calculation expression is: Among them: f j is the degree of freedom. The degree of freedom of each column is the number of levels of this column minus 1, and the total degree of freedom is equal to the sum of the degrees of freedom of each column. When F j ≤1, it can be judged that this factor has no influence on the experimental results.
[0054] As Figure 2 shown, an orthogonal experiment is carried out with the film morphology as the evaluation index. The correlation coefficient matrix between the remaining parameters after screening by range analysis and the film morphology is shown. It is found that the correlation of most parameters is not obvious, so it is necessary to further improve the analysis of variance to further judge whether the parameters are retained. The experimental method process can be referred to Figure 3 .
[0055] As a preferred embodiment, the above-mentioned multiple indicators include the curing rate, curing uniformity, and film morphology.
[0056] During the curing process, the temperature distribution is approximately determined by the energy conservation equation and the reaction heat equation, and its calculation is as follows: In the formula, ρ is the density of the curing liquid; C p is the polymerization enthalpy of the curing liquid; K is the thermal conductivity, T(r, t) is the temperature distribution at position r and time t; q_exh(r,t) is the heat generated by the exothermic reaction, and its heat can be calculated by the formula: In the formula, V r is the volume fraction of the photosensitive material in the composite matrix; ρ r is the density of the photosensitive material; ΔH r is the polymerization enthalpy of the photosensitive material; is the curing rate.
[0057] The kinetic model in UV curing can be expressed as: In the formula, E is the activation energy; s is the concentration of the photoinitiator; is the pre-exponential factor; R is the gas constant; I is the ultraviolet radiation intensity; T abs is the absolute temperature; α(r,t) is the curing conversion distribution; m and n are the reaction orders; p and q are constant exponents.
[0058] In the ultraviolet light curing polymerization reaction, the number of moles of monomer consumed (Rm) can be used to represent the photopolymerization rate, and the curing rate calculation formula is: In the formula, K p and K t are the reaction rate constants of the propagation and termination processes respectively; ψ represents the number of pairs of free radicals generated by absorbing a single photon quantum; Cm is the instantaneous monomer concentration in the polymerization reaction.
[0059] The curing degree α can be obtained by integrating the curing kinetic model over time. When α reaches 1, it indicates that the curing process is completed, and the curing time (CT) is the time required for α to reach a specific threshold. The curing speed is the change rate of the curing efficiency over time and can be obtained by differentiating the curing kinetic model.
[0060] As a preferred embodiment, the above-mentioned multiple surrogate models include the RSM model, RBF model, Kriging model, ANN model, SVM model, and / or RF model. Regarding the selection of the surrogate function of the surrogate model, the optimal kernel function of a single surrogate model can be determined by traversing the loop and comparing the root mean square error and the multiple correlation coefficient of each.
[0061] The RSM model uses second-order polynomial regression where: β 0 、β i, β ii , β ij are undetermined coefficients; X i is the i-th design variable; is a predicted value of the surrogate model; k is the variable dimension. The selection of the radial basis function of the RBF model, the correlation function γ of the Kriging model, and the activation function of the ANN model are all determined by their respective root mean square error and the multiple correlation coefficient to determine the optimal single surrogate model.
[0062] As a preferred implementation, the screening method for the surrogate model whose prediction accuracy meets the preset requirements is:
[0063] Calculate the R 2 and RMSE values between the predicted value and the true value of each surrogate model, and retain the single surrogate model with an R 2 value greater than 0.9 and an RMSE value less than 0.1;
[0064] Then, use the weighted average method to combine the retained surrogate models. The expression of the performance evaluation surrogate model obtained is where w i is the weight parameter, and its value is determined according to the R 2 weighted value corresponding to the i-th retained surrogate model; N represents the number of retained surrogate models; is the output of the i-th retained surrogate model; represents the output of the performance evaluation surrogate model. The process from the training of the surrogate model to obtaining the compromise solution set of process parameters that meets all the preset requirements of the indicators can be seen in Figure 4 .
[0065] Final predicted value As the weighted average of the predicted values of each single surrogate model, the impact on the prediction accuracy caused by the prediction differences of different surrogate models can be weakened by means of weighted average.
[0066] The combined model can effectively identify the most uncertain regions in the parameter design space, so as to add new samples at these key points to improve the accuracy of the model. Its expression is where: N is the number of surrogate models; s is the standard deviation of the predicted values of different surrogate models at a certain point or in a certain region. A large s indicates that the response values of different surrogate models in this region are very different, indicating a high degree of uncertainty or non-linearity in this region. However, a small s does not mean a high accuracy of the surrogate model in this region, but only indicates that the response values of different surrogate models in this region are close,
[0067] Therefore, the combined model can effectively identify the most uncertain regions in the design space of parameter values, and thus add new samples at these key points to improve the accuracy of the model. As a preferred implementation, S2 further includes: taking the process parameter value space corresponding to the largest output difference among the above-mentioned multiple surrogate models as the most uncertain region, and constructing a new training sample set in this most uncertain region to continue training the above-mentioned multiple surrogate models.
[0068] As a preferred implementation, in S4, a pulsed curing mode is adopted, and the UV light intensity is increased step by step and UV light irradiation is carried out intermittently to achieve curing.
[0069] In order to avoid problems such as film shrinkage and uneven heat absorption caused by long-term curing at one time and improve the uniformity of the film. As a preferred curing method, by increasing the UV light intensity step by step or gradually extending the irradiation time, the film is cured gradually from the surface to the inside to reduce internal stress; and combined with intermittent UV light irradiation to control the curing process of the film. This method is the pulsed curing mode, which can effectively reduce heat accumulation, avoid overheating and deformation of the film, and contribute to improving the density of the film. Combined with temperature control, it can effectively reduce the influence of heat on the film and prevent the film from thermally deforming.
[0070] In pulsed curing, the curing time of the curing process is an important parameter affecting the curing effect of the film. The curing time is accurately set by the control system to ensure that the photosensitive material is completely cured and avoid the influence of uncured areas on the film performance. Different curing times are set according to different curing stages to improve the overall curing effect of the film. For example, a shorter irradiation time is adopted in the initial stage to promote surface curing, and then the irradiation time is gradually extended to achieve deep curing.
[0071] As a preferred implementation, it further includes:
[0072] S5. Real-time obtain sample data including process parameter value combinations and their corresponding index values, apply the utility criterion and the infill criterion to select sample data for optimizing and improving the surrogate model whose prediction accuracy meets the preset requirements, and repeat S3.
[0073] The utility criterion and the infill criterion are introduced to maximize the model performance by effectively selecting sample points, thereby improving the prediction accuracy and optimization efficiency. The above-mentioned optimization and improvement are as follows: during the curing process, real-time data is obtained, the utility criterion and the infill criterion are applied to continuously optimize and improve the surrogate model, and finally the parameters such as the spraying speed, the distance between the UV light source and the solution, and the UV light intensity are adjusted through the feedback system until all the solution is completely cured, ensuring the uniformity and quality of the film.
[0074] The application method of the utility criterion is as follows: (1) Evaluate utility: Calculate the utility value of each candidate point using the utility criterion. (2) Select a point: Select the point with the highest utility value for actual evaluation (such as experiments, calculations). (3) Update the model: Add the new evaluation results to the dataset and update the surrogate model. (4) Iteration: Repeat step S3 until the optimization goal is achieved or the resources are exhausted.
[0075] The application method of the point addition criterion is as follows: (1) Predict uncertainty: Calculate the prediction uncertainty of the candidate point (such as standard deviation). (2) Select a point: Select the point with the greatest uncertainty or the greatest information gain according to the point addition criterion. (3) Evaluate and update: Conduct actual evaluation on the selected point, add the results to the dataset, and update the surrogate model. (4) Iteration: Repeat step S3 until the optimization goal is achieved or the resources are exhausted.
[0076] Since it is possible that the points selected by the two criteria are different, then take the union for updating the training sample data. The overall control flow can be seen in Figure 5 。
[0077] Embodiment 2
[0078] An inkjet printing UV curing film morphology regulation system, as Figure 6 shown, is used to execute an inkjet printing UV curing film morphology regulation method as described above, and includes a UV light source, an infrared interferometer, a surface morphology analyzer, a temperature sensor, a differential pressure sensor, a UV energy meter, an ultrasonic sensor, a light source control module, and a temperature control module;
[0079] Among them, the UV light source is used to provide the light required for curing; the infrared interferometer and the surface morphology analyzer are used to effectively measure and record offline the curing rate, uniformity, thickness, and morphology of the cured film to obtain the index values of various indicators; the temperature sensor, differential pressure sensor, UV energy meter, and ultrasonic sensor are used to record the process and environmental parameter data during the curing process in real time; the light source control module is used to adjust the output intensity of the UV light source according to actual needs; the temperature control module is used to monitor the temperature of the curing area in real time through the temperature sensor and perform temperature control.
[0080] As a preferred implementation method, the UV light source is an arrayed ultraviolet LED light source; the temperature control module performs temperature control through a multi-point temperature control method.
[0081] Selecting a high-intensity ultraviolet LED light source can provide sufficient energy, ensure the rapid curing of photosensitive materials, and can penetrate a relatively thick film layer to achieve deep curing. The intensity of the UV light source matching the absorption spectrum of the photosensitive material can be estimated according to the Beer-Lambert law, and its calculation formula is: In the formula, I 0 Incident ultraviolet light intensity; λc Air medium attenuation constant; Z air medium depth. At the same time, a multi-light source array or an optical light homogenizer is adopted to ensure the uniformity of UV light irradiation on the entire substrate surface, and to avoid curing differences caused by uneven light intensity.
[0082] Specifically, the UV light source includes using a high-intensity and high-density UV light source, which can effectively improve the curing efficiency. An adjustable and detachable ultraviolet LED light source is selected to ensure that the light intensity can be adjusted in real time according to different requirements in the process, and at the same time, it can be disassembled and replaced according to the UV light source wavelength matched by the absorption spectrum of different photosensitive materials. At the same time, a multi-light source array or an optical light homogenizer is adopted to ensure the uniformity of UV light irradiation on the entire substrate surface, and to avoid curing differences caused by uneven light intensity. The above uniformity can be calculated by the following formula E max is the maximum curing degree, E min is the minimum curing degree.
[0083] The infrared interferometer and the surface profiler can both effectively measure and record the curing rate, uniformity, thickness, surface topography and other quality conditions of the cured film offline. A variety of sensors, including temperature sensors, differential pressure sensors, UV energy meters and ultrasonic sensor arrays, are distributed in an array to record the process and environmental parameter data during the curing process in real time, such as light intensity monitoring, energy density, temperature, etc.
[0084] The light source control module can adjust the output intensity of the UV light source according to actual needs to ensure that the photosensitive material can obtain appropriate energy at different curing stages, and can also precisely control the illumination time of the UV light source. Specifically, it is realized by controlling different pulse inputs to achieve step-by-step curing, and to avoid film quality problems caused by a certain area being cured for a long time at one time. Then, a temperature control module is configured to monitor the temperature of the curing area in real time through a temperature sensor, and to monitor the data in real time. At the same time, for large-area substrates, multi-point temperature control is adopted to provide data support for the curing process and ensure the temperature uniformity of the entire substrate surface, and to avoid local overheating or underheating.
[0085] The related technical solutions are the same as those in Embodiment 1 and will not be elaborated here.
[0086] Generally speaking, the present invention proposes a control method and system for realizing uniform forming of the film topography through UV curing, ensuring that the film can be uniformly formed during the curing process, having a high-quality surface topography and thickness consistency, so as to solve the problems of uneven film thickness, insufficient curing rate and surface defects existing in the prior art.
[0087] Those skilled in the art can easily understand that the above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.
Claims
1. A method for controlling the morphology of an inkjet printed UV-curable film, characterized in that: include: S1. Constructing a training sample set, each training sample includes a combination of values of multiple process parameters as model input and index values of multiple indicators reflecting the curing effect as labels; the multiple process parameters are parameters that are screened and determined to have an impact on the performance of the cured film that meets preset requirements; S2. Using the training sample set, a variety of proxy models between the process parameter value combination and the index value are trained; the output of the proxy model whose prediction accuracy meets the preset requirements is weighted averaged to form a performance evaluation proxy model; S3, respectively, optimizing the indicator values of each indicator output by the performance evaluation agent model to meet the preset requirements, and determining the Pareto frontier solution set corresponding to each indicator meeting the preset requirements; intersecting each Pareto frontier solution set to obtain a process parameter selection solution set that meets the preset requirements of all indicators; S4. According to actual needs, determine the value combination of the multiple process parameters from the process parameter selection solution set to perform curing control.
2. The method for controlling the morphology of an inkjet printed UV-curable film according to claim 1, characterized in that: The method of constructing the training sample set is: The orthogonal experimental design method is used to obtain sample data through curing experiments, and the process parameters are screened based on range analysis and differential analysis. The process parameters that meet the preset requirements for the performance of the cured film are retained as the input of the proxy model. Among them, the process parameters involved in the screening include: UV power, radiation intensity, light uniformity, illumination time, light source movement speed and surface energy.
3. The method for controlling the morphology of an inkjet printed UV-curable film according to claim 1, characterized in that: The multiple indicators include curing rate, curing uniformity and film morphology.
4. The method for controlling the morphology of an inkjet printed UV-curable film according to claim 1, characterized in that: The multiple agent models include an RSM model, a RBF model, a Kriging model, an ANN model, a SVM model and / or a RF model.
5. The method for controlling the morphology of an inkjet printed UV-curable film according to claim 1, characterized in that: The method for selecting the proxy model whose prediction accuracy meets the preset requirements is as follows: Calculate the R between the predicted value and the true value of each surrogate model 2 and RMSE values, R 2 Single surrogate models with values greater than 0.9 and RMSE values less than 0.1 were retained; The retained proxy models are combined using the weighted average method, and the performance evaluation proxy model expression is obtained as follows: Among them, w i is the weight parameter, whose value is based on the R corresponding to the ith retained proxy model 2 The weighted value is determined; N represents the number of retained proxy models; is the output of the ith retained proxy model; Represents the output of the performance evaluation agent model.
6. The method for controlling the morphology of an inkjet printed UV-curable film according to claim 1, characterized in that: The S2 also includes: taking the process parameter value space corresponding to the largest output difference of the screened multiple proxy models as the most uncertain region, and constructing a new training sample set in the most uncertain region to continue training the multiple proxy models.
7. The method for controlling the morphology of an inkjet printed UV-curable film according to claim 1, characterized in that: In S4, a pulse curing mode is adopted to increase the UV light intensity step by step and perform UV light irradiation intermittently to achieve curing.
8. The method for controlling the morphology of an inkjet printed UV-curable film according to any one of claims 1 to 7, characterized in that: Also includes: S5. Acquire sample data including process parameter value combinations and their corresponding index values in real time, apply the utility criterion and the point-adding criterion to select sample data, and optimize and improve the proxy model whose prediction accuracy meets the preset requirements, and repeat S3.
9. An inkjet printing UV curing film morphology control system, characterized in that: Used to perform an inkjet printing UV curing film morphology control method as claimed in any one of claims 1 to 8, comprising a UV light source, an infrared interferometer, a surface morphology analyzer, a temperature sensor, a differential pressure sensor, a UV energy meter, an ultrasonic sensor, a light source control module and a temperature control module; Among them, the UV light source is used to provide the light required for curing; the infrared interferometer and surface morphology analyzer are used to effectively measure and record the curing rate, uniformity, thickness and morphology of the cured film offline to obtain the index values of various indicators; the temperature sensor, differential pressure sensor, UV energy meter and ultrasonic sensor are used to record the process and environmental parameter data of the curing process in real time; the light source control module is used to adjust the output intensity of the UV light source according to actual needs; the temperature control module is used to monitor the temperature of the curing area in real time through the temperature sensor and perform temperature control.
10. The inkjet printing UV curing film morphology control system according to claim 9, characterized in that: The UV light source is an arrayed ultraviolet LED light source; the temperature control module performs temperature control by means of multi-point temperature control.