Rapid calibration method and system for ultrasonic spraying process parameters based on dye tracing and image analysis

Abstract

The invention relates to an ultrasonic spraying process parameter rapid calibration method and system based on dye tracing and image analysis. The method comprises the following steps that selected dye is added into a spraying solution to serve as a tracer agent; ultrasonic spraying equipment is used for spraying on the base material according to preset parameters to form a colored coating; sampling in an effective film forming area of the coating by adopting a standard nine-point method; collecting a color image of each sampling point, and extracting and calculating a color characteristic value corresponding to the color of the tracer agent; based on the average value and the standard deviation of the characteristic values of all the sampling points, the relative thickness and the uniformity of the coating are quantitatively evaluated respectively; and through multiple groups of parameter experiments, a quantitative relation model of the process parameters, the average value and the standard deviation is established, so that an optimal process parameter window is determined. According to the method, dye tracing is adopted, and a low-cost image acquisition and processing means is combined, so that rapid, visual and quantitative calibration of ultrasonic spraying process parameters is realized, and the method is particularly suitable for rapid iterative optimization in process research, development and debugging stages.

Description

Technical Field

[0001] This invention belongs to the field of material surface treatment and manufacturing process optimization technology, specifically relating to a rapid calibration method and system for ultrasonic spraying process parameters based on dye tracing and image analysis. Background Technology

[0002] Ultrasonic spraying technology uses ultrasonic energy to atomize liquid into micron-sized droplets, which are then uniformly deposited onto the substrate surface through a nozzle. It is widely used in photovoltaic thin films, electronic packaging, biosensors, and functional coatings. The uniformity and consistency of the sprayed film thickness are key indicators determining coating performance, and these indicators are directly influenced by a complex array of process parameters, including nozzle height, moving speed, scanning spacing, liquid flow rate, and ultrasonic power.

[0003] Currently, the calibration of coating thickness and uniformity mainly relies on offline testing with precision instruments, such as profilometers and profilometers to measure film thickness, or laser confocal microscopes and ellipsometrists to analyze surface morphology and composition. While these methods offer high measurement accuracy, they generally suffer from significant drawbacks, including expensive equipment, complex and specialized operation, slow testing speed, stringent environmental requirements, and the fact that they typically only allow for destructive or sampling inspections. During the process development and debugging phases, frequent adjustments to parameter combinations and rapid evaluation of coating effects are necessary. Traditional testing methods are time-consuming, labor-intensive, and costly, severely restricting the iterative efficiency of process optimization and becoming a bottleneck in technology development. Summary of the Invention

[0004] The technical problem to be solved by this invention is to overcome the shortcomings of the prior art. In view of the urgent need for rapid parameter calibration in the research and development and debugging stage of ultrasonic spraying process, this invention provides a method and system for rapid calibration of ultrasonic spraying process parameters that is low in cost, easy to operate, provides rapid feedback, and provides intuitive quantitative results, so as to accelerate the process optimization process.

[0005] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows.

[0006] In a first aspect, the present invention provides a rapid calibration method for ultrasonic spraying process parameters based on dye tracing and image analysis, characterized by comprising the following steps:

[0007] S1. Preparation of tracer solution: Dissolve the selected dye in the spraying solvent to prepare a spraying solution containing a visual tracer;

[0008] S2, Spraying to form a film: Set the process parameters of the ultrasonic spraying equipment, and spray the solution prepared in step S1 onto the surface of a clean substrate to form a coating with tracer color.

[0009] S3. Standardized Sampling: Within the effective film-forming area of ​​the coating, at least nine analytical sampling points are determined using the nine-point sampling method;

[0010] S4. Image Acquisition and Feature Extraction: Acquire a color image of each sampling point; process each color image to extract and calculate the color feature value reflecting the tracer color intensity of that point;

[0011] S5. Quantitative evaluation of coating quality: Calculate the average value (Avg) and standard deviation (Std) of the color characteristic values ​​of all sampling points; the average value is used to evaluate the relative thickness of the coating or the amount of dye deposited, and the standard deviation is used to evaluate the uniformity of the coating thickness;

[0012] S6. Process Parameter Correlation and Optimization: Change one or more process parameters in step S2, repeat steps S2 to S5, and obtain Avg and Std values ​​corresponding to multiple sets of process parameters; establish a quantitative relationship model between process parameters and Avg and Std, and select the process parameter combination that makes the Avg value meet the thickness requirements and the Std value the smallest based on the model, and determine it as the optimal process window.

[0013] Furthermore, in step S1, the dye is methyl red. Methyl red has advantages such as chemical stability, bright color, good solubility in common solvents, and low cost. Moreover, within a certain concentration range, its color saturation shows a good correlation with the deposition amount. Specifically, it can be used as a methyl red ethanol solution.

[0014] Furthermore, in step S1, the mass concentration of methyl red in the spraying solution is 0.01% to 1%. This concentration range can ensure a significant color signal while avoiding excessive interference with the original spraying system or clogging of the nozzle.

[0015] Furthermore, in step S3, the nine-point sampling method specifically involves dividing the effective film-forming area into three equal parts along both the length and width directions, determining the center points of the four equally divided angle regions, the intersection of the four dividing lines, and the exact center point of the region, for a total of nine locations as analysis sampling points. This method can systematically evaluate the overall uniformity of the coating area.

[0016] Furthermore, in step S4, the extraction and calculation of color feature values ​​can be performed in one of two ways: First, convert the color image from the RGB color space to the HSV or HSL color space, extract the saturation channel values ​​of all pixels in the sampling point area, and calculate their average value as the color feature value of that point; Second, directly based on the RGB color space, calculate the red channel intensity value R, green channel intensity value G, and blue channel intensity value B of all pixels in the sampling point area, calculate the red saturation ratio of each pixel according to the formula S_R = R / (R + G + B) or S_R = R / (G + B), and then take the average value of the S_R values ​​of all pixels in the area as the color feature value of that point. Both methods can effectively reflect the concentration information of dye deposition.

[0017] Furthermore, in step S6, establishing the quantitative relationship model includes drawing a main effect diagram and a process window cloud map. The main effect diagram is used to visually display the influence trend of a single process parameter (such as nozzle height, line spacing, etc.) on the Avg value or Std value. The process window cloud map uses two key process parameters as the X and Y axes, and the Std value as the Z axis (three-dimensional surface) or draws contour lines (two-dimensional cloud map, where the color depth or value of the contour lines in the contour map represents the size of the Std value, and the light-colored / sparse area is the optimal window), to visualize and globally display the parameter area with the best uniformity (minimum Std value), thereby quickly locking the optimal process window.

[0018] Furthermore, the ultrasonic spraying process parameters include, but are not limited to, at least one of nozzle height, nozzle moving speed, scanning line spacing, liquid injection flow rate, and ultrasonic power.

[0019] Secondly, the present invention provides an ultrasonic spraying process parameter calibration system for performing the above method, the system comprising:

[0020] Ultrasonic spraying equipment is used to spray a solution containing tracer dye onto a substrate according to set parameters;

[0021] The image acquisition module is used to acquire color images of preset sampling points on the coating;

[0022] The image processing and analysis module is used to extract the color feature values ​​of each sampling point from the color image and calculate the average and standard deviation of the color feature values ​​of all sampling points;

[0023] The process parameter optimization module is used to establish a correlation model between process parameters and coating thickness and uniformity based on the average value and standard deviation data obtained under multiple sets of different process parameters, and output optimal process parameter suggestions.

[0024] Compared with the prior art, the present invention has the following beneficial effects:

[0025] 1. Extremely low cost: The core consumables are only ordinary dyes (such as methyl red) and common substrates (such as A4 paper and glass slides). The detection equipment can use commercial CCD cameras or smartphone cameras, combined with general image processing software or simple scripts to complete the analysis, without the need for any expensive and precision detection instruments.

[0026] 2. Simple and quick operation: No complicated sample preparation is required after spraying. Quantitative results (Avg and Std) can be obtained immediately through image acquisition and software analysis. The entire evaluation process can be completed in minutes, which greatly supports the rapid iteration and optimization of process parameters.

[0027] 3. Results visualization and quantification: The abstract concepts of "film thickness" and "uniformity" are transformed into intuitive color images and specific numerical indicators. The optimal process parameter area can be directly and clearly located through visualization tools such as main effect plots and process window cloud maps, which is highly instructive.

[0028] 4. Clear application scenarios: Specifically designed for the parameter exploration and optimization stage in the early stages of process development and debugging, it fills the market gap of lacking efficient and low-cost calibration tools in this stage, and effectively complements the subsequent online or offline quality inspection systems that use precision instruments, significantly improving R&D efficiency. Attached Figure Description

[0029] Figure 1 This is a flowchart illustrating the method of the present invention.

[0030] Figure 2 This is a schematic diagram of the nine-point sampling method used in this invention.

[0031] Figure 3 This is a schematic diagram illustrating the principle of extracting color feature values ​​through image processing.

[0032] Figure 4 This is the main effect diagram of the standard deviation of red saturation versus nozzle height and line spacing in an embodiment of the present invention.

[0033] Figure 5 This is a contour plot showing the standard deviation of red saturation versus nozzle height and line spacing in an embodiment of the present invention.

[0034] Figure 6 This is a main effect diagram of the average red saturation value versus injection volume and nozzle velocity in an embodiment of the present invention.

[0035] Figure 7 This is a main effect diagram of the standard deviation of red saturation versus injection volume and nozzle velocity in an embodiment of the present invention. Detailed Implementation

[0036] To make the technical problems solved, technical solutions, and beneficial effects of the present invention clearer, the present invention will be further described in detail below with reference to embodiments. It should be understood that the embodiments described herein are only some, not all, embodiments of the present invention, and are merely illustrative and not intended to limit the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the protection scope of the present invention.

[0037] This invention aims to solve the problems in ultrasonic spraying process development, such as the reliance on expensive and precision instruments for film thickness and uniformity calibration, cumbersome operation, and inability to quickly debug on-site. The core idea of ​​the method is to introduce a color-stable, high-contrast dye as a visual tracer into the spraying solution; after spraying, by analyzing the color information (intensity or saturation) of systematic sampling points on the coating, the deposition distribution of the dye is quantitatively characterized, thereby indirectly and quickly evaluating the film thickness (through the average eigenvalue Avg) and uniformity (through the standard deviation of eigenvalues ​​Std); finally, through multiple sets of parameter experiments, a mapping model between process parameters and Avg and Std is established, thereby scientifically determining the optimal process window.

[0038] In a first aspect, the present invention provides a rapid calibration method for ultrasonic spraying process parameters based on dye tracing and image analysis, the typical process of which is illustrated as follows: Figure 1 As shown, the process includes: preparing dyes and substrates, setting spraying parameters, spraying, nine-point sampling, saturation characterization, data recording and analysis. The analysis may include process window analysis, main effect analysis and uniformity distribution analysis.

[0039] Specifically, the rapid calibration method for ultrasonic spraying process parameters based on dye tracing and image analysis provided by this invention includes the following steps:

[0040] S1. Preparation of tracer solution: Dissolve the selected dye in the spraying solvent to prepare a spraying solution containing a visual tracer. The preferred dye is methyl red, which has good solubility in common solvents such as ethanol, acetone, and deionized water, is chemically stable, and exhibits a bright, high-contrast red color on white or light-colored substrates. Its color saturation shows a good linear relationship with the deposition amount within a certain concentration and film thickness range. The preferred mass concentration of methyl red is 0.01% to 1%.

[0041] S2. Spraying and film formation: According to the experimental design, set the various process parameters of the ultrasonic spraying equipment (such as nozzle height, moving speed, line spacing, flow rate, power, etc.), and spray the solution prepared in step S1 onto the surface of a clean substrate (such as white matte A4 paper, glass slide, etc.). After the solvent evaporates, a coating with a stable tracer color is formed.

[0042] S3. Standardized Sampling: Within the effective film-forming area of ​​the coating (using a 100mm × 100mm area centered on a 200mm × 200mm sprayed area as an example, the distribution of nine sampling points (① to ⑨) within the 100mm × 100mm effective area is shown, as follows: Figure 2 Within the area shown, nine analytical sampling points were determined using the standard nine-point sampling method. Specifically, the effective area was divided into three equal parts along its length and width. The centers of the four corner sub-regions (points ①, ②, ⑧, ⑨), the intersections of the four dividing lines (points ③, ④, ⑥, ⑦), and the exact center of the area (point ⑤) were selected, for a total of nine locations. Each sampling point area was, for example, 2mm × 2mm in size.

[0043] S4. Image Acquisition and Feature Extraction: Under standard lighting conditions, a color image of each sampling point is acquired using an image acquisition device (such as a fixed-position CCD industrial camera). The image is processed to extract color feature values. A preferred method is to convert the image from the RGB color space to the HSV color space, then extract the saturation (S) channel values ​​of all pixels within the sampling point region, and calculate the average S value of that region as the color feature value of that sampling point. This method effectively eliminates the influence of changes in lighting intensity. Figure 3 (a) and (b) exemplarily demonstrate the process of converting an RGB image to HSV space and extracting the saturation (S) channel, specifically using the red saturation of a test sample photograph as a comparison standard for uniformity and spray thickness.

[0044] S5. Quantitative Evaluation of Coating Quality: For a coating sample, calculate the arithmetic mean (Avg) and standard deviation (Std) of the color characteristic values ​​of all nine sampling points. The Avg value reflects the overall amount of dye deposited in the coating and can be used as an indicator of relative film thickness; the Std value reflects the dispersion of the deposition amount at each point. The smaller the Std value, the better the coating uniformity.

[0045] S6. Process Parameter Correlation and Optimization: Design multiple sets of experiments to systematically change one or more process parameters (e.g., using single-factor variation or multi-factor orthogonal experiments). Repeat steps S2 to S5 for each set of parameters to obtain the corresponding Avg and Std datasets. Based on these data, establish a quantitative or qualitative relationship model between process parameters and Avg and Std. Specifically, this can be achieved by drawing a main effect plot (e.g., ...). Figure 4 To analyze the single-factor influence trend, a process window cloud diagram (such as...) is plotted. Figure 5 This is used to visualize the optimal uniformity region under multiple parameter combinations. Finally, by combining the Avg value (which must meet the preset thickness requirements) and the Std value (which aims to minimize), the optimal combination of process parameters is selected, thus determining the "process window".

[0046] Secondly, the ultrasonic spraying process parameter calibration system provided by the present invention is used to automatically or semi-automatically execute the above-mentioned method, mainly including:

[0047] Ultrasonic spraying equipment is used to perform spraying tasks parametrically.

[0048] The image acquisition module may include a camera, a mounting bracket, and an illumination unit for standardized image acquisition.

[0049] The image processing and analysis module can be used on computers equipped with dedicated analysis software (such as programs written based on the OpenCV library) to automatically extract feature values ​​and calculate Avg and Std.

[0050] The process parameter optimization module can be integrated into the same software or system for data management, model building (chart generation), and optimization suggestion output.

[0051] The following specific embodiments and comparative examples further illustrate the specific implementation steps and technical effects of the present invention.

[0052] Example 1

[0053] A rapid assessment of the effect of a single process parameter (nozzle height) on coating uniformity includes the following steps:

[0054] S1. Prepare a 0.1% methyl red ethanol solution as a tracer spraying solution.

[0055] S2. Prepare clean, matte white A4 paper as the substrate. Set the ultrasonic spraying equipment to the following parameters: scanning speed 100 mm / s, line spacing 2 mm, liquid flow rate 5 mL / min, ultrasonic power 50%. Only change the nozzle height, setting it to 15 mm, 20 mm, 25 mm, 30 mm, and 35 mm respectively.

[0056] S3. Spray under each parameter to form a square coating of approximately 100 mm × 100 mm. After the solvent has completely evaporated, proceed according to... Figure 2 The nine-point sampling method shown determines nine sampling points on each coating (each point area is approximately 2 mm × 2 mm).

[0057] S4. In a standard lighting chamber, use a fixed-position 5-megapixel CCD industrial camera to capture a color image of each sampling point. All images are automatically processed using a Python script (using the OpenCV library): the image is converted from BGR to HSV color space, the saturation (S) channel values ​​of all pixels within each sampling point region are extracted, and the average S value of that region is calculated as the color feature value of that sampling point.

[0058] S5. For each coating (i.e., each group of nozzle height parameters), calculate the average (Avg) and standard deviation (Std) of its 9 sampling points. The Std value directly characterizes the spray uniformity under this parameter; the smaller the Std value, the better the uniformity.

[0059] S6. List the nozzle height with the corresponding Avg and Std values, and draw the main effect diagram.

[0060] Analysis shows that the Avg value initially increases slowly and then decreases slightly with increasing nozzle height, reflecting changes in deposition efficiency. The Std value reaches a significant minimum at a nozzle height of 25 mm. This indicates that, under current fixed parameters, a nozzle height of 25 mm achieves the best coating uniformity. This evaluation process, from experiment to results, took only a few minutes, whereas measuring film thickness at nine points using a traditional profilometer and calculating the standard deviation typically takes over two hours, demonstrating a significant efficiency difference.

[0061] Example 2

[0062] Multifactor analysis and process window determination based on experimental data, from experimental design and rapid data acquisition to graphical analysis and final determination of the process window, includes the following steps:

[0063] 1. Experimental Design: To investigate the effects of nozzle height (40, 55, 70 mm) and line spacing (10, 15, 20 mm) on uniformity (the response variable being the standard deviation of color characteristic values, Std), while controlling the infusion rate at 40 rev / min and the nozzle velocity at 15 mm / s. A full factorial design was used, comprising 9 experimental groups, as detailed in Table 1.

[0064] Table 1

[0065] Group Nozzle height (mm) Line spacing (mm) 1 40 10 2 40 15 3 40 20 4 55 10 5 55 15 6 55 20 7 70 10 8 70 15 9 70 20

[0066] 2. Data Acquisition and Calculation: Nine points were sampled for the coating after spraying each set of parameters. Images were acquired and the color feature values ​​of each point were extracted (specifically, the relative intensity value of the red channel in this embodiment). The Std values ​​of each group of 9 points were calculated, and the results are shown in Table 2.

[0067] Table 2

[0068] Group Point 1 Point 2 Point 3 Point 4 5 points Point 6 7 points 8 9 Standard deviation 1 166.10 170.10 159.97 164.37 162.55 160.49 165.17 166.10 170.10 3.30622 2 166.00 159.29 161.71 160.35 166.40 161.79 163.49 166.00 159.29 3.13669 3 152.25 151.21 134.57 133.68 154.22 154.50 145.69 158.67 151.21 9.17267 4 171.75 175.55 175.75 170.31 170.45 164.11 165.35 171.75 175.55 4.42469 5 142.89 148.41 143.04 140.71 143.81 158.09 160.56 142.89 148.41 7.62310 6 131.88 139.55 135.67 141.79 137.02 146.57 136.90 131.88 139.55 5.73555 7 158.18 156.36 155.23 158.06 159.96 156.45 158.53 158.18 156.36 1.84071 8 143.99 144.33 145.97 144.41 145.54 142.26 138.53 143.99 144.33 3.68118 9 135.62 146.29 132.78 140.76 130.51 137.19 144.29 135.62 146.29 5.27919

[0069] 3. Main Effect Analysis: Plot the main effect diagrams of nozzle height and line spacing on the Std value, such as... Figure 4 As shown in the figure, the analysis indicates that as the nozzle height increases from 40mm to 70mm, the uniformity initially worsens and then improves; as the line spacing increases from 10mm to 20mm, the uniformity continuously worsens. Furthermore, the line spacing has a greater impact on the Std value, indicating that its effect on uniformity is more significant.

[0070] 4. Process Window Cloud Map Analysis: Using nozzle height and line spacing as axes, and Std values ​​as responses, a process window cloud map is plotted, such as... Figure 5 As shown in the diagram, the cloud map clearly shows that when the nozzle height is in a lower range (<45mm) or a higher range (>65mm) and the line spacing is narrower (10-14mm), the Std value is smaller, indicating better coating uniformity. This area is the "process window" that can be quickly calibrated using the method of this invention.

[0071] 5. Spraying thickness evaluation experiment: Another experiment was designed to investigate the effects of pouring volume (20, 40, 60 rev / min) and nozzle speed (5, 15, 25 mm / s) on relative film thickness (response variable is the average value of color characteristic value Avg). The nozzle height was controlled at 55 mm and the line spacing at 15 mm. See Table 3.

[0072] Table 3

[0073] Group Infusion rate (rev / min) Nozzle speed (mm / s) 1 20 5 2 20 15 3 20 25 4 40 5 5 40 15 6 40 25 7 60 5 8 60 15 9 60 25

[0074] The main effect plot after data aggregation is as follows: Figure 6 , 7 As shown, the analysis indicates that the Avg value (representing relative thickness) is positively correlated with the injection volume and negatively correlated with the nozzle velocity, which is consistent with physical laws and verifies the rationality of the evaluation index of this invention.

[0075] Comparative Example 1

[0076] Comparative Example 1 used most of the operating steps in Example 1 for process parameter calibration. The difference was that Comparative Example 1 used a 0.1 wt% Rhodamine B ethanol solution as the dye, which was sprayed onto A4 paper to form a uniform coating and then continuously irradiated under a standard fluorescent lamp. Photos were taken at fixed sampling points at regular intervals, and their color characteristic values ​​(HSV saturation) were calculated.

[0077] Comparing Examples 1 and 2 with Comparative Example 1, the color characteristic value of the methyl red coating in Example 1 remained basically stable (fluctuation <3%) over 24 hours, while the characteristic value of the Rhodamine B coating began to show a significant decrease (>10%) after 30 minutes of irradiation, indicating that its color gradually faded under light. This demonstrates that methyl red has superior photostability, ensuring the consistency and reliability of multiple measurements during process debugging, and is more suitable as a process calibration tracer that requires repeated observation and comparison.

[0078] Comparative Example 2

[0079] Comparative Example 2 uses most of the operating steps of Example 1 for process parameter calibration. The difference is that Comparative Example 2 does not use any dyes; it only sprays a 2% (w / w) polymethyl methacrylate-acetone solution (a nearly colorless, transparent film-forming material solution). After drying, this solution forms only a very thin, transparent film on a white A4 paper substrate, causing extremely subtle changes in gloss or color depth in certain areas. It employs the same nine-point sampling and image acquisition process as the method of this invention, but due to the lack of specific color signals, only the grayscale values ​​of the image can be extracted and analyzed as feature values.

[0080] Comparing Examples 1 and 2 with Comparative Example 2, it can be found that in Comparative Example 2, the contrast changes caused by the inherent non-uniformity of the substrate, fluctuations in ambient light, and the aforementioned transparent film are extremely weak. The absolute value of the standard deviation (Std_Gray) of the calculated nine grayscale values ​​is very small, and the response to different process parameters (such as from uniform to non-uniform) is very sluggish. When the process parameters are adjusted from a uniform state to a slightly non-uniform state, the Std values ​​calculated using methyl red tracer in Examples 1 and 2 change significantly, while the Std_Gray value change of the tracer-free method is negligible, failing to effectively distinguish the quality of the process parameters. This demonstrates the crucial role of adding a high-contrast, high-stability tracer dye in achieving sensitive and reliable calibration of process parameters.

[0081] In summary, the rapid calibration method and system for ultrasonic spraying process parameters based on dye tracing and image analysis provided by this invention transforms the complex film thickness and uniformity detection into a fast and low-cost image processing problem through an ingenious "dyeing-photography-analysis" link. This significantly lowers the threshold for process development, improves optimization efficiency, and has good prospects for industrial application.

[0082] The above description is only 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 protection scope of the present invention.

Claims

1. A rapid calibration method for ultrasonic spraying process parameters based on dye tracing and image analysis, characterized in that, Includes the following steps: S1. Preparation of tracer solution: Dissolve the selected dye in the spraying solvent to prepare a spraying solution containing a visual tracer; S2, Spraying to form a film: Set the process parameters of the ultrasonic spraying equipment, and spray the solution prepared in step S1 onto the surface of a clean substrate to form a coating with tracer color. S3. Standardized Sampling: Within the effective film-forming area of ​​the coating, at least nine analytical sampling points are determined using the nine-point sampling method; S4. Image Acquisition and Feature Extraction: Acquire a color image of each sampling point; process each color image to extract and calculate the color feature value reflecting the tracer color intensity of that point; S5. Quantitative evaluation of coating quality: Calculate the average value (Avg) and standard deviation (Std) of the color characteristic values ​​of all sampling points; the average value is used to evaluate the relative thickness of the coating or the amount of dye deposited, and the standard deviation is used to evaluate the uniformity of the coating thickness; S6. Process Parameter Correlation and Optimization: Change one or more process parameters in step S2, repeat steps S2 to S5, and obtain Avg and Std values ​​corresponding to multiple sets of process parameters; establish a quantitative relationship model between process parameters and Avg and Std, and select the process parameter combination that makes the Avg value meet the thickness requirements and the Std value the smallest based on the model, and determine it as the optimal process window.

2. The rapid calibration method for ultrasonic spraying process parameters based on dye tracing and image analysis according to claim 1, characterized in that, In step S1, the dye is methyl red.

3. The rapid calibration method for ultrasonic spraying process parameters based on dye tracing and image analysis according to claim 2, characterized in that, In step S1, the mass concentration of methyl red in the spraying solution is 0.01% to 1%.

4. The rapid calibration method for ultrasonic spraying process parameters based on dye tracing and image analysis according to claim 1, characterized in that, In step S3, the nine-point sampling method specifically involves dividing the effective film-forming area into three equal parts in both length and width directions, determining the center points of the four equally divided angle regions, the intersection of the four dividing lines, and the exact center point of the region, for a total of nine locations as analysis sampling points.

5. The rapid calibration method for ultrasonic spraying process parameters based on dye tracing and image analysis according to claim 1, characterized in that, In step S4, the extraction and calculation of the color feature value reflecting the color intensity of the tracer at the point specifically involves: converting the color image from the RGB color space to the HSV or HSL color space, extracting the saturation channel values ​​of all pixels within the sampling point area, and calculating the average value of the saturation channel values ​​as the color feature value of the point.

6. The rapid calibration method for ultrasonic spraying process parameters based on dye tracing and image analysis according to claim 1, characterized in that, In step S4, the extraction and calculation of the color feature value reflecting the color intensity of the tracer at the point specifically involves: for a color image, calculating the red channel intensity value R, green channel intensity value G, and blue channel intensity value B of all pixels within the sampling point area; calculating the red saturation ratio of each pixel according to the formula S_R = R / (R + G + B) or S_R = R / (G + B); and then taking the average value of the S_R values ​​of all pixels within the area as the color feature value of the point.

7. The rapid calibration method for ultrasonic spraying process parameters based on dye tracing and image analysis according to claim 1, characterized in that, In step S6, establishing the quantitative relationship model includes drawing a main effect diagram and a process window cloud diagram; the main effect diagram is used to visually display the influence trend of a single process parameter on the Avg value or Std value; The process window cloud map uses two key process parameters as the X and Y axes, and Std values ​​as the Z axis or contour lines to visualize the parameter area with optimal uniformity.

8. The rapid calibration method for ultrasonic spraying process parameters based on dye tracing and image analysis according to claim 1, characterized in that, The ultrasonic spraying process parameters include at least one of nozzle height, nozzle moving speed, scanning line spacing, liquid injection flow rate, and ultrasonic power.

9. An ultrasonic spraying process parameter calibration system, characterized in that, The system for performing the method according to any one of claims 1 to 8 comprises: Ultrasonic spraying equipment is used to spray a solution containing tracer dye onto a substrate according to set parameters; The image acquisition module is used to acquire color images of preset sampling points on the coating; The image processing and analysis module is used to extract the color feature values ​​of each sampling point from the color image and calculate the average and standard deviation of the color feature values ​​of all sampling points; The process parameter optimization module is used to establish a correlation model between process parameters and coating thickness and uniformity based on the average value and standard deviation data obtained under multiple sets of different process parameters, and output optimal process parameter suggestions.

Images