Coating Machine Control Method and Coating Machine System

Abstract

The present invention relates to the technical field of coating control, and particularly to a coating machine control method and a coating machine system. The method includes the following steps: acquiring a coating surface image; performing surface topology analysis on the coating surface image to generate surface topology analysis data; performing surface approximation optimization on the surface topology analysis data to generate a three-dimensional model of the coating surface; performing parametric representation on the three-dimensional model of the coating surface to obtain coating surface parameter data; performing flatness analysis on the coating based on the coating surface parameter data to generate coating surface flatness data; comparing the coating surface flatness data with a preset flatness threshold to generate first flat area data, second flat area data, and third flat area data; The present invention improves the uniformity and accuracy of coating by dividing the coating terrain, compensating the coating using the coating cracking area, and predicting and comparing the coating thickness based on a neural network model.

Description

Technical Field

[0001] The present invention relates to the technical field of coating control, and particularly to a coating machine control method and a coating machine system. Background Art

[0002] In the initial stage, the control method of the coating machine mainly relied on manual operation. The operator controlled the coating thickness and uniformity by manually adjusting parameters such as the nozzle, flow rate, and speed of the coating machine. However, this method was limited by the operator's experience and skills, and there were problems such as uneven coating thickness, material waste, and low production efficiency. With the development of electronic technology, the control method of the coating machine has entered the digital and intelligent stage. Advanced control algorithms and computer vision technologies have been applied to the coating machine to achieve more precise and intelligent control. For example, using machine learning and artificial intelligence technologies, the coating machine can automatically adjust parameters according to different material properties and process requirements, optimize the coating effect, and reduce material waste. However, coating not only occurs on flat surfaces, but sometimes needs to be carried out on complex curved surfaces or irregular surfaces. In this case, the control of the coating thickness may become more difficult because it is difficult for sensors to fully cover and monitor the entire coating surface, and the viscosity and rheological properties of the coating liquid are crucial for fluid control. The viscosity of some coatings may change with temperature, pressure, or time, which may make the control of viscosity a challenge, especially when operating for a long time or under different environmental conditions, resulting in insufficient thickness uniformity of the coating. Summary of the Invention

[0003] Based on this, it is necessary to provide a coating machine control method and system to solve at least one of the above technical problems.

[0004] To achieve the above object, a coating machine control method, the method includes the following steps:

[0005] Step S1: Obtain an image of the coating surface; perform surface topology analysis on the coating surface image to generate surface topology analysis data; perform surface approximation optimization on the surface topology analysis data to generate a three-dimensional model of the coating surface; perform parametric representation on the three-dimensional model of the coating surface to obtain coating surface parameter data;

[0006] Step S2: Perform flatness analysis on the coating according to the coating surface parameter data to generate coating surface flatness data; compare the coating surface flatness data with a preset flatness threshold to generate first flat area data, second flat area data, and third flat area data; perform drying crack analysis on the first flat area data, second flat area data, and third flat area data to generate coating drying crack analysis data; perform coating compensation according to the coating drying crack analysis data to generate coating compensation data;

[0007] Step S3: Deploy surface sensors based on the coating surface parameter data and collect data to obtain real-time surface deformation data; adjust the coating jet of the coater nozzle according to the real-time surface deformation data and the coating compensation data to generate coater nozzle data; use the real-time surface deformation data and the coater nozzle data to optimize the real-time path, thereby generating real-time coating thickness adjustment data;

[0008] Step S4: Train a model with the real-time coating thickness adjustment data to generate a coating thickness prediction model; import the real-time coating thickness adjustment data into the coating thickness prediction model for thickness prediction to generate coating thickness prediction data; calculate the difference between the coating thickness prediction data and the real-time coating thickness prediction data to obtain coating thickness deviation data; compare the coating thickness deviation data with a preset coating thickness deviation threshold to generate a coater control report.

[0009] Through surface topology analysis and surface approximation, the present invention can more accurately understand the shape and quality of the coated surface. The obtained parametric representation and 3D model can be used for real-time monitoring and optimization of the coating process to ensure that the product quality meets the requirements. The generated surface model and parametric representation provide in-depth understanding of the coating process, which helps to improve process parameters and operations. The coating effect can be optimized by adjusting the parameters, reducing waste and the defective rate. The 3D model and parametric representation of the coated surface can be used to support product design and process planning, helping to design a product surface that better meets the requirements. By analyzing the surface topology analysis data and parametric representation, data-driven decisions are made to improve production efficiency and quality management level. Through flatness analysis and crack analysis, quality problems on the coated surface can be detected and solved in advance, improving the consistency and flatness of the product surface. Through detailed analysis of different flatness regions, targeted coating compensation can be carried out to reduce the defective rate and production cost. Using the coating compensation data, the automated system can adjust the coating process in real time, improving production efficiency and reducing the need for manual intervention. Through detailed analysis and adjustment of the coated surface, more consistent product surface performance can be achieved, improving the overall product quality. Using surface sensors to collect surface deformation data in real time enables the system to quickly understand the state of the coated surface and make timely adjustments. Using the real-time deformation data and compensation data, the coating machine nozzle is adjusted to enable it to adapt to surface changes in a timely and precise manner, ensuring coating quality. Based on the real-time deformation data and nozzle data, path optimization is carried out to ensure uniform coating thickness during the coating process, improving product quality. Through real-time adjustment, unnecessary coating errors are avoided, production efficiency is improved, material waste is reduced, and costs are lowered. By training a model with the real-time coating thickness adjustment data, a coating thickness prediction model is established. This model can help the system better understand the actual situation and provide more accurate predictions in future coating processes. Using the trained model, the real-time coating thickness adjustment data is imported into the model to achieve real-time prediction of the coating thickness. This helps to detect potential problems in a timely manner and take control measures to ensure product quality. By calculating the difference between the coating thickness prediction data and the real-time coating thickness data, coating thickness deviation data is obtained. Such deviation data can intuitively reflect the difference between the actual coating situation and the expectation. Comparing the coating thickness deviation data with a preset coating thickness deviation threshold generates a coating machine control report. Such a report can be used for real-time monitoring and triggering an alarm or automatic adjustment when the preset threshold is exceeded, thus keeping the coating quality within a reasonable range. Therefore, the present invention improves the uniformity and accuracy of coating by dividing the coating terrain, compensating the coating using the coated crack area, and predicting and comparing the coating thickness based on a neural network model.

[0010] The beneficial effects of the present invention are as follows: Through surface topology analysis and surface approximation optimization, a three-dimensional surface model is generated and parametrically represented. This provides the basic surface feature data for subsequent steps. Using the coating surface parameter data for flatness analysis, data of different flat regions are further generated. In addition, through crack analysis, the coating quality can be more comprehensively evaluated, providing a basis for subsequent coating compensation. Based on the surface parameter data, sensor deployment is carried out to collect real-time deformation data. This real-time data collection can help monitor the dynamic changes during the coating process, thereby better understanding the actual situation. Using the crack analysis data, coating compensation is performed in step S2 to generate coating compensation data. In step S3, the real-time deformation data and compensation data are used for nozzle adjustment and path optimization to ensure the uniformity and quality of the coating. Through model training on the real-time coating thickness adjustment data, a coating thickness prediction model is generated. This can be used to predict future coating thicknesses. By comparing the predicted data with the actual coating data, coating thickness deviation data is generated and compared with a preset threshold to form a coating machine control report. Therefore, the present invention improves the uniformity and accuracy of coating by dividing the coating terrain, compensating the coating using the coating dry crack area, and predicting and comparing the coating thickness based on a neural network model. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 is a schematic flow chart of the steps of a coating machine control method;

[0012] Figure 2 is Figure 1 a detailed implementation step flow chart of step S2 in

[0013] Figure 3 is Figure 2 a detailed implementation step flow chart of step S23 in

[0014] Figure 4 is Figure 1 a detailed implementation step flow chart of step S3 in

[0015] The realization, functional features, and advantages of the object of the present invention will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] The technical method of the present invention for patents will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of them. All other embodiments obtained by those skilled in the art within the scope of the present invention without creative efforts belong to the scope of protection of the present invention.

[0017] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus repeated descriptions thereof will be omitted. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor methods and / or microcontroller methods.

[0018] It should be understood that although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0019] To achieve the above object, please refer to Figures 1 to 4 , a coating machine control method, the method comprising the following steps:

[0020] Step S1: Obtain a coating surface image; perform surface topology analysis on the coating surface image to generate surface topology analysis data; perform surface approximation optimization on the surface topology analysis data to generate a three-dimensional model of the coating surface; perform parametric representation on the three-dimensional model of the coating surface to obtain coating surface parameter data;

[0021] Step S2: Perform flatness analysis on the coating based on the coating surface parameter data to generate coating surface flatness data; compare the coating surface flatness data with a preset flatness threshold to generate first flat area data, second flat area data, and third flat area data; perform drying crack analysis on the first flat area data, second flat area data, and third flat area data to generate coating drying crack analysis data; perform coating compensation based on the coating drying crack analysis data to generate coating compensation data;

[0022] Step S3: Deploy surface sensors based on the coating surface parameter data and collect data to obtain surface real-time deformation data; adjust the coating jet of the coating machine nozzle according to the surface real-time deformation data and the coating compensation data to generate coating machine nozzle data; perform real-time path optimization using the surface real-time deformation data and the coating machine nozzle data to generate real-time coating thickness adjustment data;

[0023] Step S4: Perform model training on the real-time coating thickness adjustment data to generate a coating thickness prediction model; import the real-time coating thickness adjustment data into the coating thickness prediction model for thickness prediction to generate coating thickness prediction data; calculate the difference between the coating thickness prediction data and the real-time coating thickness prediction data to obtain coating thickness deviation data; compare the coating thickness deviation data with a preset coating thickness deviation threshold to generate a coater control report.

[0024] Through surface topology analysis and surface approximation, the present invention can more accurately understand the shape and quality of the coated surface. The obtained parametric representation and 3D model can be used for real-time monitoring and optimization of the coating process to ensure that the product quality meets the requirements. The generated surface model and parametric representation provide in-depth understanding of the coating process, which helps to improve process parameters and operations. The coating effect can be optimized by adjusting the parameters, reducing waste and the defective rate. The 3D model and parametric representation of the coated surface can be used to support product design and process planning, helping to design product surfaces that better meet the requirements. By analyzing the surface topology analysis data and parametric representation, data-driven decisions are made to improve production efficiency and quality management level. Through flatness analysis and crack analysis, quality problems on the coated surface can be discovered and solved in advance, improving the consistency and flatness of the product surface. Through detailed analysis of different flatness regions, targeted coating compensation can be carried out to reduce the defective rate and production costs. Using the coating compensation data, the automated system can adjust the coating process in real time, improving production efficiency and reducing the need for manual intervention. Through detailed analysis and adjustment of the coated surface, more consistent product surface performance can be achieved, improving the overall product quality. Using surface sensors to collect surface deformation data in real time enables the system to quickly understand the state of the coated surface and make timely adjustments. Using the real-time deformation data and compensation data, the coating machine nozzle is adjusted to enable it to promptly and precisely adapt to surface changes, ensuring coating quality. Based on the real-time deformation data and nozzle data, path optimization is carried out to ensure uniform coating thickness during the coating process, improving product quality. Through real-time adjustment, unnecessary coating errors are avoided, production efficiency is improved, material waste is reduced, and costs are lowered. By training a model with the real-time coating thickness adjustment data, a coating thickness prediction model is established. This model can help the system better understand the actual situation and provide more accurate predictions during future coating processes. Using the trained model, the real-time coating thickness adjustment data is imported into the model to achieve real-time prediction of the coating thickness. This helps to promptly discover potential problems and take control measures to ensure product quality. By calculating the difference between the coating thickness prediction data and the real-time coating thickness data, coating thickness deviation data is obtained. This deviation data can intuitively reflect the difference between the actual coating situation and the expectation. By comparing the coating thickness deviation data with a preset coating thickness deviation threshold, a coating machine control report is generated. Such a report can be used for real-time monitoring and trigger an alarm or automatic adjustment when the preset threshold is exceeded, thus keeping the coating quality within a reasonable range. Therefore, the present invention improves the uniformity and accuracy of coating by dividing the coating terrain, compensating the coating using the coated crack area, and predicting and comparing the coating thickness based on a neural network model.

[0025] In an embodiment of the present invention, with reference to Figure 1As shown, it is a schematic diagram of the step flow of a coating machine control method according to the present invention. In this embodiment, the coating machine control method includes the following steps:

[0026] Step S1: Obtain the coating surface image; perform surface topology analysis on the coating surface image to generate surface topology analysis data; perform surface approximation optimization on the surface topology analysis data to generate a three-dimensional model of the coating surface; perform parametric representation on the three-dimensional model of the coating surface to obtain the coating surface parameter data;

[0027] In an embodiment of the present invention, the coating surface is imaged by using a high-resolution camera or sensor. Ensure the clarity and accuracy of the image for subsequent analysis. Preprocess the image, such as denoising, smoothing, etc., to prepare for surface topology analysis. Apply computer vision or image processing algorithms, such as edge detection, feature extraction, etc., to obtain surface topology data. This may involve extracting contours, feature points, or topological structure information from the image. Use mathematical modeling methods, such as point cloud-based surface approximation techniques, to process the data obtained from the surface topology analysis to generate a three-dimensional model of the coating surface. This may involve surface fitting algorithms, such as least squares fitting or Bezier curve fitting, to optimize the data and generate a three-dimensional model. Parametrically represent the generated three-dimensional model, possibly using methods such as parametric surfaces, B-spline surfaces, etc., to effectively describe the characteristics of the coating surface. Extract the required surface parameter data, such as curvature, concavity, surface features, etc., from the parametric surface model. This step may include the application of mathematical formulations or specific algorithms to obtain the required parametric data.

[0028] Step S2: Perform flatness analysis on the coating according to the coating surface parameter data to generate coating surface flatness data; compare the coating surface flatness data with a preset flatness threshold to generate first flat area data, second flat area data, and third flat area data; perform dry crack analysis on the first flat area data, second flat area data, and third flat area data to generate coating dry crack analysis data; perform coating compensation according to the coating dry crack analysis data to generate coating compensation data;

[0029] In the embodiments of the present invention, flatness analysis is performed by using coating surface parameter data. This may involve mathematical calculations of the surface curvature model, such as indicators like the local inclination of the surface and elevation changes, to describe the flatness of the surface. The data obtained from the flatness analysis is compared with a preset flatness threshold to divide the surface into different flat regions. It may include a first flat region (a region that meets the flatness standard), a second flat region (a region close to the flatness standard), and a third flat region (a region that does not meet the flatness standard). Dry crack analysis is performed on the divided different flat regions. This may include quantitative analysis of the crack density, crack length, crack orientation, etc. in each flat region. Using image processing and analysis techniques, methods such as threshold segmentation of cracks and edge detection may need to be considered. According to the dry crack analysis data, the areas and degrees that need coating compensation are determined. This may involve some algorithms, such as increasing the coating amount in areas with more cracks to improve the uniformity of coating. Coating compensation data is generated, which can be a set of parameters describing the additional coating amount or other adjustment parameters that should be applied in different regions.

[0030] Step S3: Deploy surface sensors based on the coating surface parameter data and collect data to obtain real-time surface deformation data; adjust the coating spray of the coating machine nozzle according to the real-time surface deformation data and the coating compensation data to generate coating machine nozzle data; use the real-time surface deformation data and the coating machine nozzle data to perform real-time path optimization, thereby generating real-time coating thickness adjustment data;

[0031] In the embodiments of the present invention, according to the coating surface parameter data, an appropriate type of sensor (such as an optical sensor, a pressure sensor, or a laser scanner) is selected and deployed beside the coating machine or on the workpiece surface. The sensor should be able to obtain real-time surface deformation data. The sensor acquisition system is started to continuously collect and record the surface deformation data. This may involve the installation of data acquisition equipment, sensor calibration, and the processing of data streams. The real-time surface deformation data obtained is compared and analyzed with the previously generated coating compensation data. According to these data, the spraying mode and coating amount of the coating machine nozzle are adjusted. A control system is established to input the real-time deformation data collected by the sensor and adjust the parameters of the coating machine nozzle in real time according to the coating compensation data. This can be an automated system or be monitored and adjusted in real time by an operator. Using the real-time surface deformation data and the adjusted coating machine nozzle data, a real-time path optimization algorithm is executed. This may include dynamically adjusting the movement path of the coating machine arm or nozzle according to the characteristics of the surface deformation to achieve more uniform coating. Combining the real-time path optimization results, the thickness that needs to be adjusted in the coating area is determined. According to the optimized path and the coating machine nozzle adjustment data, real-time coating thickness adjustment data is generated.

[0032] Step S4: Perform model training on the real-time coating thickness adjustment data to generate a coating thickness prediction model; import the real-time coating thickness adjustment data into the coating thickness prediction model for thickness prediction to generate coating thickness prediction data; calculate the difference between the coating thickness prediction data and the real-time coating thickness prediction data to obtain coating thickness deviation data; compare the coating thickness deviation data with a preset coating thickness deviation threshold to generate a coater control report.

[0033] In the embodiment of the present invention, the real-time coating thickness adjustment data is organized into a training set. Each sample should include input features (such as surface real-time deformation data, coater nozzle data, etc.) and corresponding output labels (actual coating thickness adjustment data). Select an appropriate machine learning or deep learning model for training, such as a neural network, regression model, etc., to establish a coating thickness prediction model. Use the training set to train the model and adjust the model parameters to minimize the gap between the predicted value and the actual value. Input the real-time coating thickness adjustment data into the trained model to generate coating thickness prediction data. Calculate the difference between the coating thickness prediction data and the actual coating thickness adjustment data to obtain coating thickness deviation data. Analyze the coating thickness deviation data to understand the differences and changes in the coating process. Compare the coating thickness deviation data with a preset coating thickness deviation threshold. Generate a coater control report according to the comparison result, indicating whether it is within the allowable range. If not, the report may include specific deviation values and recommended correction measures.

[0034] Preferably, step S1 includes the following steps:

[0035] Step S11: Obtain a coating surface image using laser scanning;

[0036] Step S12: Perform image point cloud filtering on the coating surface image to generate a coating point cloud image; perform image preprocessing on the coating point cloud image to generate a standard coating point cloud image, where the image preprocessing includes image brightness enhancement, image geometric pixel change, and image edge recognition;

[0037] Step S13: Extract surface features from the coating point cloud image to generate surface feature description data; perform surface topology analysis based on the surface feature description data to generate surface topology analysis data; perform triangular mesh conversion on the surface topology analysis data according to the grid modeling technology to generate a coating surface triangular mesh model;

[0038] Step S14: Perform surface approximation optimization on the coating surface triangular mesh model to generate a three-dimensional coating surface model; perform parametric representation on the three-dimensional coating surface model to obtain coating surface parameter data.

[0039] The present invention scans the coated surface using laser scanning technology to obtain image data of the surface. The laser scanning provides high-precision coated surface data, providing an accurate basis for subsequent processing. The obtained coated surface image from laser scanning is subjected to point cloud filtering to convert it into point cloud data. Then, preprocessing is performed on the point cloud image, including enhancing image brightness, adjusting geometric pixel changes, and performing edge recognition, etc., to generate a standard coated point cloud image. The point cloud filtering and image preprocessing improve the quality and accuracy of the data, making it better for subsequent surface feature extraction and topological analysis. Surface feature extraction is performed on the coated point cloud image to identify and extract the feature information of the coated surface. Based on the surface feature description data, surface topological analysis is carried out to generate surface topological analysis data. Then, the grid modeling technology is used to convert the surface topological analysis data into a triangular grid model of the coated surface. The surface feature extraction and topological analysis provide a detailed description and data analysis of the coated surface, generating a structured triangular grid model, providing a basis for subsequent surface approximation and parametric representation. By optimizing the surface approximation of the triangular grid model of the coated surface, a more accurate three-dimensional model of the coated surface is generated. Parametric representation is performed on the three-dimensional model of the coated surface, representing the surface as a set of parametric data. The surface approximation optimization improves the accuracy and smoothness of the coated surface model, and the parametric representation enables the coated surface to be more conveniently further analyzed and processed, providing a basis for subsequent tasks.

[0040] In an embodiment of the present invention, the coated surface is scanned using a laser scanner to generate an image containing surface geometric information. The coated surface image is converted into point cloud data. Point cloud filtering technology can be used to remove noise and irrelevant information to obtain a clear coated point cloud image. Image preprocessing is performed on the coated point cloud image, including brightness enhancement to improve image quality, geometric pixel changes to adjust the image shape, and edge recognition to extract the edge information of the surface, so as to generate a standard coated point cloud image. Surface feature extraction is performed on the coated point cloud image to identify and extract the feature information of the surface. Feature extraction algorithms such as normal calculation and curvature calculation can be used. Based on the extracted surface feature description data, surface topological analysis is carried out to analyze the connection relationship and topological structure of the surface, generating surface topological analysis data. The grid modeling technology is used to convert the surface topological analysis data into a triangular grid model. This process divides the surface into many small triangles to form a triangular grid model of the coated surface. Surface approximation optimization is performed on the triangular grid model of the coated surface. Through techniques such as fitting, the surface is made smoother and more accurate, generating an optimized three-dimensional model of the coated surface. Parametric representation is performed on the three-dimensional model of the coated surface, representing the surface as a set of parametric data. Common parametric methods include texture mapping, curve parameterization, etc. In this way, it is convenient to further analyze, edit, and process the coated surface, and obtain the parametric data of the coated surface.

[0041] Preferably, step S2 includes the following steps:

[0042] Step S21: Perform flatness analysis on the coating according to the coating surface parameter data to generate coating surface flatness data; compare the coating surface flatness data with a preset flatness threshold. When the coating surface flatness data is greater than the preset flatness threshold, perform the first flat area division based on the coating surface flatness data to obtain the first flat area data; when the coating surface flatness data is equal to the preset flatness threshold, perform the second flat area division based on the coating surface flatness data to obtain the second flat area data; when the coating surface flatness data is less than the preset flatness threshold, perform the third flat area division based on the coating surface flatness data to obtain the third flat area data;

[0043] Step S22: Mark the third flat area data with a high coating priority to generate a high-priority coating area; mark the second flat area data with a medium priority to generate a medium-priority coating area; mark the first flat area data with a low priority to generate a low-priority coating area;

[0044] Step S23: Analyze the adjacent area coating rheological characteristics of the coating curved surface three-dimensional model through the low-priority coating area, medium-priority coating area, and high-priority coating area to generate coating rheological characteristic analysis data; perform drying crack analysis on the low-priority coating area, medium-priority coating area, and high-priority coating area based on the coating rheological characteristic analysis data to generate coating drying crack analysis data;

[0045] Step S24: Perform coating compensation on the low-priority coating area, medium-priority coating area, and high-priority coating area based on the coating drying crack analysis data to generate coating compensation data.

[0046] The present invention obtains an image of the coated surface through laser scanning, which can provide high-precision data for subsequent processing and analysis. Point cloud filtering and preprocessing are performed on the image of the coated surface, including brightness enhancement, geometric pixel variation, and edge recognition, which can improve the image quality and eliminate noise and interference. Through the extraction of surface feature description data and topological analysis, a detailed feature analysis of the coated surface can be carried out to help understand its structure and properties. The grid modeling technology is applied to convert the surface topological analysis data into a triangular mesh model, which helps further geometric modeling and analysis. The surface approximation optimization is performed on the three-dimensional model of the coated surface to generate a more accurate surface model and perform parametric representation to obtain the parametric data of the coated surface. By analyzing and comparing the flat data of the coated surface, the division of the flat area is implemented, the flatness of the coated surface can be evaluated, and the coated area data with different priorities can be obtained. By analyzing the rheological properties and drying cracks of the coated area, the coating behavior in different areas and the risk of crack formation can be understood, providing a basis for subsequent coating optimization and compensation. Based on the drying crack analysis data, coating compensation is performed on the low-priority coated area, medium-priority coated area, and high-priority coated area, which can improve the coating quality and reduce the occurrence of cracks and defects.

[0047] As an example of the present invention, refer to Figure 2 shown, in this example, step S2 includes:

[0048] Step S21: Perform flatness analysis on the coating according to the coated surface parameter data to generate the flat data of the coated surface; compare the flat data of the coated surface with a preset flat threshold. When the flat data of the coated surface is greater than the preset flat threshold, the first flat area division is performed based on the flat data of the coated surface to obtain the first flat area data; when the flat data of the coated surface is equal to the preset flat threshold, the second flat area division is performed based on the flat data of the coated surface to obtain the second flat area data; when the flat data of the coated surface is less than the preset flat threshold, the third flat area division is performed based on the flat data of the coated surface to obtain the third flat area data;

[0049] In the embodiments of the present invention, parameter data of the coated surface, such as height or curvature, is obtained by using a suitable measurement method (such as laser scanning or sensors). Flatness analysis is performed on the parameter data of the coated surface, and the flatness or surface change of the coated surface is evaluated through calculation or statistical methods. According to the requirements and objectives of the coating, an appropriate flatness threshold is set as the judgment criterion. This threshold can be a fixed value determined based on experience or can be set according to product specifications and requirements. The flatness data of the coated surface is compared with the preset flatness threshold. By comparing the magnitude relationship between the flatness data and the threshold, the flatness degree of the coated surface is determined. If the flatness data of the coated surface is greater than the preset flatness threshold, a first flat region is divided based on the flatness data of the coated surface. The specific division method may involve using algorithms or rules to determine the boundaries and ranges of the regions. If the flatness data of the coated surface is equal to the preset flatness threshold, a second flat region is divided based on the flatness data of the coated surface. Similarly, the boundaries and ranges of the regions are determined according to algorithms or rules. If the flatness data of the coated surface is less than the preset flatness threshold, a third flat region is divided based on the flatness data of the coated surface. Similarly, the boundaries and ranges of the regions are determined according to algorithms or rules. According to the divided first, second, and third flat regions, corresponding region data is generated, which can be images of the regions, boundary coordinates, or other data describing the characteristics of the flat regions.

[0050] Step S22: Mark the data of the third flat region with high coating priority to generate a high-priority coating region; mark the data of the second flat region with medium priority to generate a medium-priority coating region; mark the data of the first flat region with low priority to generate a low-priority coating region;

[0051] In the embodiments of the present invention, the boundaries of the third flat region are determined by dividing according to the flatness data of the coated surface. For each point or pixel within the third flat region, it is marked as a high-priority coating region. Specific marking values or colors can be used to represent the high-priority region. The boundaries of the second flat region are determined by dividing according to the flatness data of the coated surface. For each point or pixel within the second flat region, it is marked as a medium-priority coating region. Similarly, specific marking values or colors can be used to represent the medium-priority region. The boundaries of the first flat region are determined by dividing according to the flatness data of the coated surface. For each point or pixel within the first flat region, it is marked as a low-priority coating region. Similarly, specific marking values or colors can be used to represent the low-priority region.

[0052] Step S23: Analyze the coating rheological properties of adjacent regions of the three-dimensional coating surface model through the low-priority coating region, medium-priority coating region, and high-priority coating region to generate coating rheological property analysis data; based on the coating rheological property analysis data, conduct dry crack analysis on the low-priority coating region, medium-priority coating region, and high-priority coating region to generate coating dry crack analysis data;

[0053] In the embodiment of the present invention, by connecting and importing the three-dimensional coating surface model to the coating rheological property analysis software, the coating surface is divided into a low-priority coating region, a medium-priority coating region, and a high-priority coating region, and rheological property analysis is performed on each region. This analysis can involve parameters such as the rheological properties of the coating material, surface tension, coating speed, etc. During the analysis, considering the priorities of different regions, different rheological property parameters can be adopted for the low-priority region, medium-priority region, and high-priority region. Based on the data obtained from the coating rheological property analysis, dry crack analysis is carried out, and the prediction and analysis of dry cracks are respectively performed on the low-priority coating region, medium-priority coating region, and high-priority coating region. The dry crack analysis software or algorithm can be used to predict the possibility and degree of cracks in different regions according to factors such as rheological property data, drying conditions, and the coating surface model. The analysis results can include information such as the number of cracks, the length of cracks, and the depth of cracks.

[0054] Step S24: Based on the coating dry crack analysis data, perform coating compensation on the low-priority coating region, medium-priority coating region, and high-priority coating region to generate coating compensation data.

[0055] In the embodiment of the present invention, by synthesizing and analyzing the coating dry crack analysis data, these data can include information such as the position, size, and shape of cracks, as well as data related to coating parameters (such as coating speed, coating thickness, etc.) and drying conditions (such as temperature, humidity, etc.). According to the analysis data, determine the compensation strategies for the low-priority coating region, medium-priority coating region, and high-priority coating region. The compensation strategies can include adjusting coating parameters (such as speed, pressure, etc.) or changing the composition, viscosity, etc. of the coating material. According to the determined compensation strategies, perform coating compensation on the low-priority coating region, medium-priority coating region, and high-priority coating region. Use relevant coating compensation techniques or algorithms to calculate the required compensation parameters or adjustment values according to the crack analysis data and compensation strategies. These compensation data can include adjusted coating speed, coating amount, coating angle, etc.

[0056] Preferably, step S23 includes the following steps:

[0057] Step S231: Conduct first-priority uniform coating simulation on the high-priority coating area to obtain high-priority coating area data; conduct second-priority uniform coating simulation on the medium-priority coating area to obtain medium-priority coating area data; conduct regional curvature analysis on the high-priority coating area data to generate concave area curvature data; based on the concave area curvature data, conduct adjacent area coating rheological property analysis on the high-priority coating area data and the medium-priority coating area data to generate first flow rate data;

[0058] Step S232: Conduct third-priority uniform coating simulation on the low-priority coating area to obtain low-priority coating area data; conduct regional curvature analysis on the low-priority coating area data to generate convex area curvature data; based on the convex area curvature data, conduct adjacent area coating rheological property analysis on the low-priority coating area data and the medium-priority coating area data to generate second flow rate data;

[0059] Step S233: Based on the concave area curvature data and the convex area curvature data, conduct adjacent area coating rheological property analysis on the low-priority coating area and the high-priority coating area to generate third flow rate data;

[0060] Step S234: Integrate the first flow rate data, the second flow rate data, and the third flow rate data to generate coating rheological property analysis data; according to the coating rheological property analysis data, conduct simulated drying and shaping treatment on the low-priority coating area, the medium-priority coating area, and the high-priority coating area to obtain coating area drying data;

[0061] Step S235: Conduct coating crack analysis on the coating area drying data through the coating drying crack analysis formula to generate coating drying crack analysis data.

[0062] The present invention obtains coating area data by uniformly coating and simulating high, medium, and low priority coating areas. This helps to ensure the uniformity of the coating area and avoid undercoating or overcoating. The regional curvature of the first priority coating area is analyzed to generate concave area curvature data; the regional curvature of the third priority coating area is analyzed to generate convex area curvature data. These curvature data provide information on the surface morphology of the area and contribute to subsequent coating rheological property analysis. Based on the concave area curvature data and the convex area curvature data, the coating rheological property analysis of adjacent areas is carried out. By analyzing the rheological properties between coating areas, the rate and mode of coating flow can be determined to further optimize the coating process and ensure coating consistency. According to the coating rheological property analysis data, the low, medium, and high priority coating areas are subjected to simulated drying and shaping treatment to obtain coating area drying data. This helps to evaluate the performance and effect of the coating during the drying process and provides a basis for subsequent crack analysis. Through the coating drying crack analysis formula, the coating area drying data is subjected to crack analysis to generate coating drying crack analysis data. By analyzing the formation reasons and distribution of cracks, the degree of cracks can be understood, providing a basis for designing a coating compensation scheme.

[0063] As an example of the present invention, refer to Figure 3 shown, in this example, step S23 includes:

[0064] Step S231: Perform a first priority uniform coating simulation on the high priority coating area to obtain first priority coating area data; perform a second priority uniform coating simulation on the medium priority coating area to obtain second priority coating area data; analyze the regional curvature of the first priority coating area data to generate concave area curvature data; based on the concave area curvature data, perform the coating rheological property analysis of adjacent areas on the first priority coating area data and the second priority coating area data to generate first flow rate data;

[0065] In the embodiments of the present invention, by performing uniform coating simulation on the high-priority coating area, coating data for the first-priority coating area is obtained. This involves determining the spraying method of the coating agent, coating process parameters, and accurate control of the spraying equipment to ensure uniformity during the coating process. A similar coating simulation process is carried out on the medium-priority coating area to obtain coating data for the second-priority coating area. The purpose of this step is to determine the coating characteristics of the medium-priority coating area for subsequent rheological property analysis. The curvature analysis method is used to analyze the surface of the first-priority coating area to obtain the curvature data of the concave area. These data can be used to identify areas where depressions are likely to occur. Using the coating rheological property analysis method and combining with the curvature data of the concave area, the flow rate analysis of adjacent areas of the first-priority coating area and the second-priority coating area is carried out. This step aims to determine the fluidity and coating uniformity of the coating material in these areas.

[0066] Step S232: Perform third-priority uniform coating simulation on the low-priority coating area to obtain data for the third-priority coating area; perform regional curvature analysis on the data for the third-priority coating area to generate curvature data for the convex area; based on the curvature data for the convex area, perform adjacent-region coating rheological property analysis on the data for the third-priority coating area and the data for the second-priority coating area to generate second flow rate data;

[0067] In the embodiments of the present invention, by performing third-priority uniform coating simulation on the low-priority coating area. Appropriate coating process parameters and equipment control are used to ensure the uniformity of the coating process in this area. Similar to the previous steps, regional curvature analysis is performed on the surface of the third-priority coating area. By analyzing the curvature data, the convex area, that is, the area with a larger curvature, can be determined. Using the curvature data for the convex area, adjacent-region coating rheological property analysis of the third-priority coating area and the second-priority coating area is carried out. This analysis aims to determine the fluidity and coating uniformity of the coating material in these areas. According to the results of the adjacent-region coating rheological property analysis, second flow rate data is generated. These data represent the flow rate situation of the coating material between the third-priority coating area and the second-priority coating area.

[0068] Step S233: Based on the curvature data for the concave area and the curvature data for the convex area, perform adjacent-region coating rheological property analysis on the low-priority coating area and the high-priority coating area to generate third flow rate data;

[0069] In the embodiments of the present invention, curvature data of low-priority coating areas and high-priority coating areas is obtained. The curvature data of the concave area represents the area with a smaller curvature, while the curvature data of the convex area represents the area with a larger curvature. The curvature data of the concave area and the curvature data of the convex area are applied to the low-priority coating area and the high-priority coating area to analyze the coating rheological properties of adjacent areas. This analysis aims to determine the fluidity and coating uniformity of the coating material in adjacent areas. According to the results of the coating rheological property analysis of adjacent areas, third flow rate data is generated. These data represent the flow rate of the coating material between the low-priority coating area and the high-priority coating area.

[0070] Step S234: Integrate the first flow rate data, the second flow rate data, and the third flow rate data to generate coating rheological property analysis data; perform simulated drying and shaping processing on the low-priority coating area, the medium-priority coating area, and the high-priority coating area according to the coating rheological property analysis data to obtain coating area drying data;

[0071] In the embodiments of the present invention, the first flow rate data, the second flow rate data, and the third flow rate data are integrated. This can be accomplished through statistical methods, averaging, or other corresponding data processing methods. The purpose is to obtain comprehensive coating rheological property analysis data, including flow rate information of coating areas with different priorities. Use the integrated coating rheological property analysis data to perform simulated drying and shaping processing on the low-priority coating area, the medium-priority coating area, and the high-priority coating area. This step can be simulated and analyzed using a computer model or professional software. According to specific requirements, set parameters for the drying and shaping process, such as temperature, humidity, time, etc., and use the coating rheological property analysis data to establish corresponding models for each coating area. These models can consider the characteristics and properties of different coating areas. Generally, this involves numerical calculation and simulation methods, running the models to simulate the drying and shaping process of the low-priority coating area, the medium-priority coating area, and the high-priority coating area. According to the set processing parameters and models, calculate the changes and properties of the coating area during the drying and shaping process. Through the simulated drying and shaping process, obtain the drying data of the coating area. These data can include indicators such as the drying time, drying temperature, and drying quality of the coating area.

[0072] Step S235: Analyze the coating cracks of the coating area drying data through the coating drying crack analysis formula to generate coating drying crack analysis data.

[0073] In the embodiments of the present invention, by obtaining the drying data of the coating area from the previous step S234, including indicators such as the drying time, drying temperature, and drying quality of the coating area. The coating drying crack analysis formula can be selected according to specific applications and requirements. These formulas are obtained through experiments and theoretical analyses and are used to predict and evaluate the possible crack conditions during the coating drying process. Apply the selected coating drying crack analysis formula to the drying data of the coating area. According to the requirements of the formula, use the drying data as input variables to calculate the possible crack conditions in the coating area. Generate coating drying crack analysis data based on the results of applying the coating drying crack analysis formula. These data can include crack indicators such as the position, length, width, and density of the cracks.

[0074] Preferably, the coating drying crack analysis formula in step S235 is specifically as follows:

[0075]

[0076] In the formula, C(x, y) represents the coating drying crack analysis data, indicating the crack degree at a specific position (x, y) in the coating area, x represents the horizontal position coordinate of the coating area, y represents the vertical position coordinate of the coating area, t represents the time of the drying process, h represents the thickness of the coating layer, represents the second-order partial derivative of the coating layer thickness function with respect to the horizontal position x, indicating the curvature of the coating layer in the horizontal direction, represents the second-order partial derivative of the coating layer thickness function with respect to the vertical position y, indicating the curvature of the coating layer in the vertical direction, represents the third-order mixed partial derivative of the coating layer thickness function with respect to x, y, and t, indicating the change rate of the coating layer at different positions and time points, and μ represents the abnormal adjustment value of the coating drying crack analysis.

[0077] The present invention constructs a coating drying crack analysis formula. The thickness of the coating layer in the formula is one of the important factors determining the coating quality and crack formation. By observing the change in the coating layer thickness, the non-uniformity and potential crack risk during the coating process can be evaluated. A functional relationship is formed based on the mutual relationship between the horizontal position coordinate and the vertical position coordinate of the coating area and the above parameters:

[0078]

[0079] By adjusting the coating layer thickness function and the second-order partial derivatives with respect to the position coordinates. These partial derivatives represent the curvature of the coating layer in the horizontal and vertical directions. By analyzing the change in curvature, the unevenness of the coating layer and the potential crack formation area can be detected. The coating layer thickness function The third-order mixed partial derivative with respect to position coordinates and time. This mixed partial derivative represents the rate of change of the coating layer at different positions and time points. By analyzing the rate of change, the dynamic changes and the trend of crack formation during the coating process can be detected. The position coordinates of the coating area are used to locate the specific position for analysis, enabling a quantitative assessment of the crack degree within the coating area. The time variable represents the time of the drying process. This variable allows for a dynamic analysis of the crack degree during the drying process to understand the evolution of cracks. The abnormal adjustment value μ for coating drying crack analysis is used to correct the errors and deviations caused by the complexity and non-ideality of the actual system. It can correct the difference between the theoretical assumptions in the formula and the actual system, improve the accuracy and reliability of coating drying crack analysis, and more accurately generate the crack degree C(x, y) at a specific position (x, y) in the coating area. At the same time, parameters such as the horizontal position coordinate and the time of the drying process in the formula can be adjusted according to the actual situation, thus adapting to different coating drying crack analysis scenarios and improving the applicability and flexibility of the algorithm. When using the conventional coating drying crack analysis formula in the art, the crack degree at a specific position (x, y) in the coating area can be obtained. By applying the coating drying crack analysis formula provided by the present invention, the crack degree at a specific position (x, y) in the coating area can be calculated more precisely. By analyzing the interactions between the above parameters and variables, the formation and development of cracks during the coating process can be quantitatively evaluated. This helps to identify the causes of crack formation and take corresponding measures to improve the coating process, optimize the coating quality, and reduce the occurrence of cracks.

[0080] Preferably, step S3 includes the following steps:

[0081] Step S31: Deploy surface sensors based on the coating surface parameter data and collect data to obtain real-time surface deformation data;

[0082] Step S32: Adjust the coating spray of the coating machine nozzle according to the real-time surface deformation data and the coating compensation data to generate coating machine nozzle data;

[0083] Step S33: Plan the coating path for the coating machine nozzle data through the coating area dynamic programming formula to generate coating area path planning data;

[0084] Step S34: Control the coating equipment parameters for the low-priority coating area, medium-priority coating area, and high-priority coating area by using the coating area path planning data and the real-time surface deformation data to generate area coating thickness adjustment data;

[0085] Step S35: Perform real-time path optimization based on the coating thickness adjustment data and the coating area path planning data, so as to generate real-time coating thickness adjustment data.

[0086] Through the deployment and data acquisition of the surface sensor, the present invention can obtain the deformation data of the coating surface in real time. This provides accurate monitoring and evaluation of the surface state during the coating process for the adjustment and optimization of subsequent steps. According to the real-time deformation data and the pre-determined coating compensation data, the coating machine nozzle is adjusted and controlled. This enables the jet coating process to be adjusted according to the real-time surface deformation, improving the coating uniformity and quality. The coating area dynamic planning formula performs path planning on the coating machine nozzle data, considering factors such as coating targets, nozzle positions, and movement limitations. This enables the coating process to perform path planning according to actual requirements, improving the coating efficiency and consistency. According to the coating area path planning and the real-time surface deformation data, the coating equipment parameters, such as coating speed, nozzle spacing, etc., are adjusted to achieve precise control of the coating thickness in different coating areas. This helps to meet the coating requirements of different areas and improve the coating quality and consistency. Combining the coating thickness adjustment data and the coating area path planning data, real-time path optimization is performed. This can further improve the consistency and control accuracy of the coating thickness during the coating process, and improve the coating efficiency and quality.

[0087] As an example of the present invention, refer to Figure 4 As shown, in this example, step S3 includes:

[0088] Step S31: Deploy the surface sensor based on the coating surface parameter data and perform data acquisition to obtain the real-time surface deformation data;

[0089] In the embodiments of the present invention, a sensor suitable for measuring surface deformation is selected. Commonly used surface deformation sensors include fiber optic sensors, resistance strain gauges, piezoelectric sensors, etc. According to specific coating application requirements and surface deformation characteristics, an appropriate sensor type is selected. According to the coating area and surface deformation conditions, a deployment plan for the sensor is designed. The number, position, and arrangement method of the sensor need to be considered. Usually, sensors are installed at different positions or key parts of the coating area to obtain comprehensive surface deformation data. The sensor is connected to a data acquisition system and necessary calibration is carried out. Ensure that the sensor works properly and can accurately measure surface deformation data. Start the data acquisition system and begin to collect surface deformation data in real time. According to the preset sampling frequency, the deformation conditions of the coating surface are regularly recorded. These data can be converted into electrical signals or digital signals by the sensor and then processed and saved by the data acquisition system. The collected surface deformation data is transmitted to the corresponding algorithm or control system for processing and analysis. Feedback regulation is carried out in real time according to the surface deformation data, such as the coating spray adjustment for the coating machine nozzle in step S32.

[0090] Step S32: Adjust the coating spray of the coating machine nozzle according to the real-time surface deformation data and coating compensation data to generate coating machine nozzle data;

[0091] In the embodiments of the present invention, the real-time surface deformation data and coating compensation data are obtained. The real-time surface deformation data is measured in step S31, and the coating compensation data may come from previous calibration experiments or model predictions. The real-time surface deformation data and coating compensation data are analyzed and processed to extract relevant information about the adjustment of the coating machine nozzle. This may include the magnitude and direction of the deformation, as well as the adjustment amount required for coating compensation. Based on the analyzed and processed data, an algorithm for nozzle adjustment is designed and implemented. These algorithms can adjust parameters such as the spray speed, spray angle, and nozzle position of the coating machine nozzle according to the requirements of the coating process to achieve compensation and optimization during the coating process. According to the results calculated by the nozzle adjustment algorithm, nozzle data suitable for the coating machine is generated. These nozzle data can include information such as the setting of the spray speed, the adjustment of the spray angle, and the movement of the nozzle position to achieve precise control of the coating process. The generated nozzle data is transmitted to the coating machine control system and applied to the coating spray process. The coating machine control system adjusts the operation of the nozzle according to the requirements of the nozzle data to achieve precise control and compensation during the coating process.

[0092] Step S33: Perform coating path planning on the coating machine nozzle data through the coating area dynamic programming formula to generate coating area path planning data;

[0093] In the embodiments of the present invention, the coating area is defined. This can be achieved by setting the boundaries of the coating target or using a known coating template. The coating area can be a region on a two-dimensional plane or a specific area on the surface of the coating object. According to the coating requirements and the characteristics of the coating area, a suitable dynamic programming formula is designed. This formula should take into account the position and movement constraints of the coating machine nozzle, as well as the optimization objectives of coating quality and efficiency. Based on the designed dynamic programming formula, a coating path planning algorithm is implemented. This algorithm can utilize the idea of dynamic programming and obtain the optimal path and coating parameters for each point through iterative calculations. Apply the path planning data generated by the coating area path planning algorithm to the coating machine nozzle data. This includes determining the position of the coating machine nozzle and coating parameters, such as spraying speed, spraying angle, etc., at each moment. Transmit the generated coating area path planning data to the coating machine control system and apply them during the coating process. The coating machine control system will control the movement and parameters of the nozzle according to the requirements of the path planning data to achieve path planning within the coating area.

[0094] Step S34: Use the coating area path planning data and the real-time surface deformation data to control the coating equipment parameters for the low-priority coating area, medium-priority coating area, and high-priority coating area, and generate regional coating thickness adjustment data;

[0095] In the embodiments of the present invention, the path planning data of the coating area is obtained. This may include information such as the coordinates, shape, and size of each area. Determine the priority of each area, i.e., low, medium, and high priorities. Use devices such as sensors or cameras to monitor the deformation data of the coating surface in real time. These deformation data can include information such as the shape change and thickness change of the surface. Develop a device parameter control strategy to deal with low, medium, and high-priority coating areas. For low-priority areas, it may be necessary to reduce the coating speed or increase the coating amount to ensure full coverage. For high-priority areas, it may be necessary to increase the coating speed or reduce the coating amount to obtain a more uniform coating. Integrate the path planning data and the deformation data to form a comprehensive coating control data set, and conduct real-time data analysis to understand the actual status and effects of each area during the coating process. According to the real-time analysis results, generate coating thickness adjustment data, which may involve setting the target coating thickness for each area and calculating the amount of parameters that need to be adjusted. Feed the generated coating thickness adjustment data back to the coating equipment to adjust the coating parameters in real time, which can be achieved through an automatic control system to ensure real-time performance and accuracy.

[0096] Step S35: Conduct real-time path optimization through the coating thickness adjustment data and the coating area path planning data, thereby generating real-time coating thickness adjustment data.

[0097] In the embodiments of the present invention, by developing or selecting appropriate algorithms for real-time path optimization based on coating thickness adjustment data and path planning data, methods such as dynamic path planning, optimization algorithms, or machine learning models may be involved. The previously generated coating thickness adjustment data is combined with the path planning data of the coating area to ensure that the path data of the coating area and the coating thickness data to be adjusted can be effectively corresponding and matched. During the coating process, the performance of the coating machine or system and the coating situation are monitored in real time, and the designed real-time path optimization algorithm is used to adjust the coating path according to the real-time data. The adjusted path may involve parameters such as the movement trajectory, speed, and angle of the coating head. Based on the optimized path, real-time coating thickness adjustment data is regenerated, which may include adjustments to parameters such as coating amount, coating speed, and coating angle. The real-time generated coating thickness adjustment data is fed back to the coating equipment to achieve real-time adjustment, ensuring that the system can continuously receive and process the real-time generated data to achieve dynamic adjustment of the coating process.

[0098] Preferably, the dynamic programming formula for the coating area in step S33 is specifically as follows:

[0099]

[0100] In the formula, J(u) represents the performance index of the coating path planning, u represents the control variable vector of the coating path planning, which includes the control parameters of the coating machine nozzle at different times and positions, T represents the total time of the coating process, L(r, r, u) represents the local cost function in the coating path planning, which is the local cost term related to the position r, speed r, and control variable u during the coating process, Ω represents the spatial domain of the coating area, represents the global cost function in the coating path planning, which is the global cost term related to the position r, control variable u, and the gradient of the coating agent on the coating area related, and ε represents the dynamic programming anomaly correction amount of the coating area.

[0101] The present invention constructs a dynamic programming formula for the coating area. J(u) in the formula is the performance index of the path planning, which is used to measure the total cost of the coating process. By minimizing the cost function, an optimized coating path planning can be achieved. Minimizing the cost function can bring various beneficial effects, such as reducing the usage amount of the coating agent, improving the uniformity of the coating quality, and reducing the coating time. According to the mutual relationship between the control variable vector of the coating path planning and the above parameters, a functional relationship is formed:

[0102]

[0103] By adjusting the control variable vector u, the control parameters of the coater nozzle at different times and positions are included. By selecting appropriate control variables, the coating path and nozzle movement can be affected. By adjusting the control variables, the following beneficial effects can be achieved: Optimal path planning: Selecting appropriate control variables can optimize the coating path, such as avoiding obstacles and reducing backtracking, thereby improving the coating efficiency and quality. Coating agent distribution control: The control variables can affect the distribution of the coating agent on the coating area. By adjusting the spraying position and intensity of the coating agent, uniform distribution of the coating agent and reduction of waste can be achieved. Coating speed and thickness control: The control variables can also adjust the speed and spraying thickness of the coater, thereby achieving control of the coating speed and coating layer thickness. Is expressed as the local cost function in the coating path planning, which is the local cost item related to the position r, speed And the control variable u during the coating process. It can be designed according to specific requirements, considering various factors. By designing an appropriate local cost function, the following beneficial effects can be achieved: Obstacle avoidance and collision avoidance: By incorporating the obstacle avoidance cost into the local cost function, the path planning algorithm can avoid obstacles and collisions with other objects. Path smoothing and stability: By considering the change rates of speed and position, the path planning can be made smoother and more stable, reducing the jitter and instability of the movement. The global cost function in the coating path planning It is related to the position r, the control variable u, and the gradient of the coating agent on the coating area during the coating process Relevant global cost items. It reflects the distribution of the coating agent on the coating area and the uniformity of the coating quality. By designing an appropriate global cost function, the following beneficial effects can be achieved: Uniform distribution of the coating agent: By considering the gradient of the coating agent on the coating area, the coating agent can be evenly distributed over the entire area, thereby improving the uniformity of the coating quality and reducing the waste of the coating agent. Optimization of the coating quality: The global cost function can be used to optimize the coating quality, such as minimizing the concentration change of the coating agent and reducing the mixing of the coating agent. The abnormal correction amount ε of the coating area dynamic programming is used to correct the errors and deviations caused by the complexity and non-ideality of the actual system. It can correct the difference between the theoretical assumptions in the formula and the actual system, improve the accuracy and reliability of the coating area dynamic programming, generate the performance index J(u) of the coating path planning more accurately. At the same time, parameters such as the local cost function and the global cost function in the coating path planning can be adjusted according to the actual situation, so as to adapt to different coating area dynamic programming scenarios, and improve the applicability and flexibility of the algorithm. When using the conventional coating area dynamic programming formula in the art, the performance index of the coating path planning can be obtained. By applying the coating area dynamic programming formula provided by the present invention, the performance index of the coating path planning can be calculated more accurately. The coating area dynamic programming formula realizes the optimization of the coating path planning by minimizing the cost function and adjusting the control variables, designing the local cost function and the global cost function. Through reasonable path planning and control strategies, beneficial effects of reducing costs, improving coating efficiency and quality can be achieved.

[0104] Preferably, step S34 includes the following steps:

[0105] Step S341: Obtain initial coating motion data based on the coating machine nozzle data, where the initial coating motion data includes coating pressure data and coating speed data; Use the coating area path planning data to confirm the center points of the low-priority coating area, medium-priority coating area, and high-priority coating area, and generate coating area center point data;

[0106] Step S342: Perform area confirmation on the coating area center point data through infrared spectroscopy technology. When it is determined that the coating area center point data is a low-priority coating area, reduce the coating pressure data and coating speed data according to the surface real-time deformation data to obtain low-priority coating data;

[0107] Step S343: When it is determined that the center point data of the coating area is the medium-priority coating area, the coating pressure data and the coating speed data are stabilized according to the surface real-time deformation data to obtain the medium-priority coating data; when it is determined that the center point data of the coating area is the high-priority coating area, the coating pressure data and the coating speed data are increased according to the surface real-time deformation data to obtain the high-priority coating data;

[0108] Step S344: Integrate the low-priority coating data, the medium-priority coating data, and the high-priority coating data to generate the priority area coating data; perform a coating adsorption uniformity analysis on the priority area coating data and the coating rheological property analysis data to obtain the coating adsorption uniformity data;

[0109] Step S345: Control the coating equipment parameters of the coating machine nozzle according to the coating adsorption uniformity data to generate the area coating thickness adjustment data.

[0110] The present invention obtains the initial coating motion data through the nozzle data, including the coating pressure and the coating speed data, providing the basic data for the subsequent steps. Using the path planning data to confirm the center points of the low, medium, and high-priority coating areas helps to accurately locate different areas and provides the basis for subsequent processing. Confirming the center points of the coating areas through infrared spectroscopy technology, combined with the surface real-time deformation data, the coating pressure and the coating speed are processed according to different priority areas. The low-priority area is reduced, the medium-priority area is stabilized, and the high-priority area is increased. This processing method makes the coating amount more uniform in different areas. Integrating the low, medium, and high-priority coating data to generate the priority area coating data and performing a coating adsorption uniformity analysis on these data helps to evaluate the coating uniformity, thus ensuring that the final coating effect is more consistent. According to the coating adsorption uniformity data, controlling the coating equipment parameters of the coating machine nozzle, this precise control helps to adjust the coating thickness and ensure that the coating amount in different areas meets the design requirements. Combining the control of the coating equipment parameters, the final area coating thickness adjustment data is generated, providing the optimal parameter settings for the coating process.

[0111] In the embodiments of the present invention, by installing sensors near the nozzles of the coater, the working data of the coater are collected in real time, including coating pressure and coating speed. Using a path planning algorithm, the coating area is divided into low, medium, and high-priority coating areas, and the central point coordinates of each area are calculated to generate the central point data of the coating area. Using infrared spectroscopy technology, the central point data of the coating area are verified and confirmed as low-priority coating areas. The real-time surface deformation data are obtained, and the coating pressure and coating speed in the low-priority coating areas are reduced through a control algorithm to ensure more uniform coating in these areas. It is determined that the central point data of the coating area are of medium priority, and the medium coating pressure and coating speed are maintained to ensure uniform coating in these areas. The low, medium, and high-priority coating data are integrated into one to form the coating data of the priority area. The coating data are analyzed using a coating adsorption uniformity analysis tool to ensure overall uniformity. According to the coating adsorption uniformity data, the parameters of the coater, such as coating pressure, speed, etc., are adjusted. According to the parameter adjustment, the final regional coating thickness adjustment data are generated to meet the product quality and coating requirements.

[0112] Preferably, step S4 includes the following steps:

[0113] Step S41: Collect historical data of the real-time coating thickness adjustment data to obtain historical coating effect data; perform data preprocessing on the historical coating effect data to generate standard historical coating effect data, where the data preprocessing includes data cleaning, filling in missing data values, and data standardization;

[0114] Step S42: Divide the standard historical coating effect data into data sets to generate a model training set and a model test set; use a neural network model to train the model training set to generate a coating thickness prediction model; optimize the coating thickness prediction model through the model test set to generate a coating thickness prediction model;

[0115] Step S43: Import the real-time coating thickness adjustment data into the coating thickness prediction model for thickness prediction to generate coating thickness prediction data; calculate the difference between the coating thickness prediction data and the real-time coating thickness prediction data to obtain coating thickness deviation data;

[0116] Step S44: Compare the coating thickness deviation data with a preset coating thickness deviation threshold. When the coating thickness deviation data are greater than the coating thickness deviation threshold, control optimization processing is performed on the coater nozzles; when the coating thickness deviation data are less than or equal to the coating thickness deviation threshold, a coater control report is generated based on visualization technology.

[0117] The present invention helps to establish a comprehensive understanding of past coating performance by collecting historical coating effect data, providing a rich data source for subsequent model training. Through preprocessing steps such as cleaning, filling missing values, and standardization, the quality and consistency of historical coating effect data are ensured, improving the accuracy and stability of subsequent models. A neural network model is used to train the historical data to generate a coating thickness prediction model. Through the optimization of the model test set, the model's generalization ability on unseen data is ensured. The model is applied to real-time data to achieve real-time prediction of coating thickness, improving the response speed and accuracy of the coating process. When the coating thickness deviation exceeds the set threshold, the parameters are adjusted in real time by optimizing the control of the coating machine nozzle, making the coating process more stable and meeting the requirements. Visualization technology is used to generate a coating machine control report, providing intuitive data display and analysis results, providing decision-making support for operators, and at the same time realizing the traceability and continuous improvement of the coating process.

[0118] In an embodiment of the present invention, by using a data acquisition system to record in real time the adjustment data during the coating process, including coating machine operation parameters, environmental conditions, etc., the real-time data is stored in a database to establish a historical data set. Data cleaning is performed to handle outliers, anomalies, etc., and appropriate algorithms are used to fill in missing values to ensure data integrity. Data standardization is carried out to bring the data to the same scale so that the neural network can learn better. The standard historical coating effect data is divided into a training set and a test set, usually by random division or division by time. An appropriate neural network architecture is selected, such as a multi-layer perceptron (MLP) or a recurrent neural network (RNN), to process time series data. The model is trained using the training set, and the model parameters are adjusted to minimize the prediction error. The test set is used to evaluate the model performance, and the model is adjusted and optimized according to the evaluation results, such as adjusting the learning rate, increasing the model complexity, etc. The trained neural network model is embedded in the control system of the coating machine to obtain real-time coating thickness prediction data and make predictions based on the real-time data. The difference between the real-time coating thickness data and the prediction data is calculated to obtain the coating thickness deviation data.

[0119] In this specification, a coating machine control system is provided for implementing the coating machine control method described above. The coating machine control system includes:

[0120] A three-dimensional reconstruction module for acquiring a coating surface image; performing surface topology analysis on the coating surface image to generate surface topology analysis data; performing surface approximation optimization on the surface topology analysis data to generate a three-dimensional model of the coating surface; and performing parametric representation on the three-dimensional model of the coating surface to obtain coating surface parameter data.

[0121] The coating compensation module is used to perform flatness analysis on the coating according to the coating surface parameter data to generate coating surface flatness data; compare the coating surface flatness data with a preset flatness threshold to generate first flat area data, second flat area data, and third flat area data; perform dry crack analysis on the first flat area data, second flat area data, and third flat area data to generate coating dry crack analysis data; perform coating compensation according to the coating dry crack analysis data to generate coating compensation data;

[0122] The path optimization module is used to deploy surface sensors based on the coating surface parameter data and collect data, so as to obtain surface real-time deformation data; adjust the coating jet of the coating machine nozzle according to the surface real-time deformation data and the coating compensation data to generate coating machine nozzle data; use the surface real-time deformation data and the coating machine nozzle data to perform real-time path optimization, so as to generate real-time coating thickness adjustment data;

[0123] The deviation comparison module is used to perform model training on the real-time coating thickness adjustment data to generate a coating thickness prediction model; import the real-time coating thickness adjustment data into the coating thickness prediction model for thickness prediction to generate coating thickness prediction data; calculate the difference between the coating thickness prediction data and the real-time coating thickness prediction data to obtain coating thickness deviation data; compare the coating thickness deviation data with a preset coating thickness deviation threshold to generate a coating machine control report.

[0124] The beneficial effects of the present invention are as follows: Through surface topology analysis and surface approximation optimization, a three-dimensional surface model is generated and parametrically represented. This provides basic surface feature data for subsequent steps. Using the coating surface parameter data for flatness analysis further generates data for different flat areas. In addition, through crack analysis, the coating quality can be more comprehensively evaluated, providing a basis for subsequent coating compensation. Based on the surface parameter data, sensors are deployed to collect real-time deformation data. This real-time data collection can help monitor the dynamic changes during the coating process, thus better understanding the actual situation. Using the crack analysis data, coating compensation is performed in step S2 to generate coating compensation data. In step S3, the real-time deformation data and the compensation data are used to adjust the nozzle and optimize the path to ensure the uniformity and quality of the coating. Through model training on the real-time coating thickness adjustment data, a coating thickness prediction model is generated. This can be used to predict future coating thicknesses. By comparing the predicted data with the actual coating data, coating thickness deviation data is generated and compared with a preset threshold to form a coating machine control report. Therefore, the present invention improves the uniformity and accuracy of the coating by dividing the coating terrain, compensating the coating using the coating dry crack area, and predicting and comparing the coating thickness based on a neural network model.

[0125] Therefore, from any perspective, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the application documents are intended to be encompassed within the present invention.

[0126] The above description is only a specific implementation manner of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather to the broadest scope consistent with the principles and novel features invented herein.

Claims

1. A coating machine control method, characterized in that, it includes the following steps: Step S1: Obtain the coating surface image; perform surface topology analysis on the coating surface image to generate surface topology analysis data; perform surface approximation optimization on the surface topology analysis data to generate a three-dimensional model of the coating surface; perform parametric representation on the three-dimensional model of the coating surface to obtain coating surface parameter data; Step S2: Perform flatness analysis on the coating according to the coating surface parameter data to generate coating surface flatness data; compare the coating surface flatness data with a preset flatness threshold to generate first flat area data, second flat area data, and third flat area data; Perform dry crack analysis on the first flat area data, second flat area data, and third flat area data to generate coating dry crack analysis data; perform coating compensation according to the coating dry crack analysis data to generate coating compensation data; Step S3: Deploy surface sensors based on the coating surface parameter data and collect data to obtain surface real-time deformation data; Adjust the coating spray of the coating machine nozzle according to the surface real-time deformation data and the coating compensation data to generate coating machine nozzle data; Use the surface real-time deformation data and the coating machine nozzle data to perform real-time path optimization to generate real-time coating thickness adjustment data; Step S3 includes the following steps: Step S31: Deploy surface sensors based on the coating surface parameter data and collect data to obtain surface real-time deformation data; Step S32: Adjust the coating spray of the coating machine nozzle according to the surface real-time deformation data and the coating compensation data to generate coating machine nozzle data; Step S33: Plan the coating path for the coating machine nozzle data through the coating area dynamic programming formula to generate coating area path planning data; the coating area dynamic programming formula in Step S33 is as follows: In the formula, represents the performance index for coating path planning, represents the control variable vector for coating path planning, which contains the control parameters of the coating machine nozzle at different times and positions, represents the total time of the coating process, represents the local cost function in coating path planning, which is the local cost item related to the position , speed and control variable in the coating process, represents the spatial domain of the coating area, represents the global cost function in coating path planning, which is the global cost item related to the position , control variable and the gradient of the coating agent on the coating area in the coating process, represents the abnormal correction amount of coating area dynamic programming; Step S34: Control the coating equipment parameters for the low-priority coating area, medium-priority coating area, and high-priority coating area by using the coating area path planning data and the surface real-time deformation data to generate area coating thickness adjustment data; Step S34 includes the following steps: Step S341: Obtain initial coating motion data based on the coating machine nozzle data, where the initial coating motion data includes coating pressure data and coating speed data; use the coating area path planning data to confirm the center points of the low-priority coating area, medium-priority coating area, and high-priority coating area to generate coating area center point data; Step S342: Confirm the area through the infrared spectroscopy technology for the coating area center point data. When it is determined that the coating area center point data is the low-priority coating area, reduce the coating pressure data and coating speed data according to the surface real-time deformation data to obtain low-priority coating data; Step S343: When it is determined that the center point data of the coating area is the medium-priority coating area, the coating pressure data and the coating speed data are stabilized according to the real-time surface deformation data to obtain the medium-priority coating data; when it is determined that the center point data of the coating area is the high-priority coating area, the coating pressure data and the coating speed data are increased according to the real-time surface deformation data to obtain the high-priority coating data; Step S344: Integrate the low-priority coating data, the medium-priority coating data, and the high-priority coating data to generate the priority area coating data; perform a coating adsorption uniformity analysis on the priority area coating data and the coating rheological characteristic analysis data to obtain the coating adsorption uniformity data; Step S345: Control the coating equipment parameters of the coating machine nozzle according to the coating adsorption uniformity data to generate the regional coating thickness adjustment data; Step S35: Perform real-time path optimization through the coating thickness adjustment data and the coating area path planning data to generate the real-time coating thickness adjustment data; Step S4: Perform model training on the real-time coating thickness adjustment data to generate a coating thickness prediction model; import the real-time coating thickness adjustment data into the coating thickness prediction model for thickness prediction to generate coating thickness prediction data; calculate the difference between the coating thickness prediction data and the real-time coating thickness prediction data to obtain the coating thickness deviation data; compare the coating thickness deviation data with the preset coating thickness deviation threshold to generate a coating machine control report.

2. The coating machine control method according to claim 1, wherein, Step S1 includes the following steps: Step S11: Use laser scanning to obtain the coating surface image; Step S12: Perform image point cloud filtering on the coating surface image to generate a coating point cloud image; perform image preprocessing on the coating point cloud image to generate a standard coating point cloud image, where the image preprocessing includes image brightness enhancement, image geometric pixel change, and image edge recognition; Step S13: Extract surface features from the coating point cloud image to generate surface feature description data; perform surface topology analysis based on the surface feature description data to generate surface topology analysis data; perform triangular mesh conversion on the surface topology analysis data according to the grid modeling technology to generate a coating surface triangular mesh model; Step S14: Perform surface approximation optimization on the coating surface triangular mesh model to generate a three-dimensional coating surface model; perform parametric representation on the three-dimensional coating surface model to obtain the coating surface parameter data.

3. The coating machine control method according to claim 1, wherein, Step S2 includes the following steps: Step S21: Perform flatness analysis on the coating based on the coating surface parameter data to generate coating surface flatness data; compare the coating surface flatness data with a preset flatness threshold. When the coating surface flatness data is greater than the preset flatness threshold, perform the first flat area division based on the coating surface flatness data to obtain the first flat area data; when the coating surface flatness data is equal to the preset flatness threshold, perform the second flat area division based on the coating surface flatness data to obtain the second flat area data; when the coating surface flatness data is less than the preset flatness threshold, perform the third flat area division based on the coating surface flatness data to obtain the third flat area data; Step S22: Mark the third flat area data with high coating priority to generate a high-priority coating area; mark the second flat area data with medium priority to generate a medium-priority coating area; mark the first flat area data with low priority to generate a low-priority coating area; Step S23: Analyze the adjacent area coating rheological characteristics of the coating curved surface three-dimensional model through the low-priority coating area, medium-priority coating area, and high-priority coating area to generate coating rheological characteristic analysis data; perform drying crack analysis on the low-priority coating area, medium-priority coating area, and high-priority coating area based on the coating rheological characteristic analysis data to generate coating drying crack analysis data; Step S24: Perform coating compensation on the low-priority coating area, medium-priority coating area, and high-priority coating area based on the coating drying crack analysis data to generate coating compensation data.

4. The coating machine control method according to claim 3, wherein, Step S23 includes the following steps: Step S231: Perform a first-priority uniform coating simulation on the high-priority coating area to obtain first-priority coating area data; perform a second-priority uniform coating simulation on the medium-priority coating area to obtain second-priority coating area data; perform area curvature analysis on the first-priority coating area data to generate concave area curvature data; perform adjacent area coating rheological characteristic analysis on the first-priority coating area data and the second-priority coating area data based on the concave area curvature data to generate the first flow rate data; Step S232: Perform a third-priority uniform coating simulation on the low-priority coating area to obtain third-priority coating area data; perform area curvature analysis on the third-priority coating area data to generate convex area curvature data; perform adjacent area coating rheological characteristic analysis on the third-priority coating area data and the second-priority coating area data based on the convex area curvature data to generate the second flow rate data; Step S233: Perform adjacent area coating rheological characteristic analysis on the low-priority coating area and the high-priority coating area based on the concave area curvature data and the convex area curvature data to generate the third flow rate data; Step S234: Integrate the first flow rate data, the second flow rate data, and the third flow rate data to generate coating rheological characteristic analysis data; perform simulated drying and shaping processing on the low-priority coating area, the medium-priority coating area, and the high-priority coating area according to the coating rheological characteristic analysis data to obtain coating area drying data; Step S235: Perform coating crack analysis on the coating area drying data through the coating drying crack analysis formula to generate coating drying crack analysis data.

5. The coating machine control method according to claim 4, characterized in that, the coating drying crack analysis formula in step S235 is as follows: In the formula, represents the coating drying crack analysis data, indicating the crack degree at a specific position in the coating area ; represents the horizontal position coordinate of the coating area, represents the vertical position coordinate of the coating area, represents the time of the drying process, represents the thickness of the coating layer, represents the second-order partial derivative of the coating layer thickness function with respect to the horizontal position , indicating the curvature of the coating layer in the horizontal direction, represents the second-order partial derivative of the coating layer thickness function with respect to the vertical position , indicating the curvature of the coating layer in the vertical direction, represents the coating layer thickness function with respect to , and the third-order mixed partial derivative, indicating the change rate of the coating layer at different positions and time points, represents the abnormal adjustment value of the coating drying crack analysis.

6. The coating machine control method according to claim 1, characterized in that, step S4 includes the following steps: Step S41: Collect historical data of the real-time coating thickness adjustment data to obtain historical coating effect data; perform data preprocessing on the historical coating effect data to generate standard historical coating effect data, where the data preprocessing includes data cleaning, data missing value filling, and data standardization; Step S42: Divide the standard historical coating effect data into a data set to generate a model training set and a model test set; use a neural network model to train the model training set to generate a coating thickness prediction model; optimize the coating thickness prediction model through the model test set to generate a coating thickness prediction model; Step S43: Import the real-time coating thickness adjustment data into the coating thickness prediction model for thickness prediction to generate coating thickness prediction data; calculate the difference between the coating thickness prediction data and the real-time coating thickness prediction data to obtain coating thickness deviation data; Step S44: Compare the coating thickness deviation data with a preset coating thickness deviation threshold. When the coating thickness deviation data is greater than the coating thickness deviation threshold, perform control optimization processing on the coating machine nozzle; when the coating thickness deviation data is less than or equal to the coating thickness deviation threshold, generate a coating machine control report based on visualization technology.

7. A coating machine control system, characterized in that, for executing the coating machine control method according to claim 1, the coating machine control system includes: A three-dimensional reconstruction module, configured to obtain a coating surface image; perform surface topology analysis on the coating surface image to generate surface topology analysis data; perform surface approximation optimization on the surface topology analysis data to generate a three-dimensional coating surface model; perform parametric representation on the three-dimensional coating surface model to obtain coating surface parameter data; A coating compensation module, configured to perform flatness analysis on the coating according to the coating surface parameter data to generate coating surface flatness data; compare the coating surface flatness data with a preset flatness threshold to generate first flat area data, second flat area data, and third flat area data; perform drying crack analysis on the first flat area data, the second flat area data, and the third flat area data to generate coating drying crack analysis data; perform coating compensation according to the coating drying crack analysis data to generate coating compensation data; A path optimization module is used to deploy surface sensors based on coating surface parameter data and collect data, so as to obtain real-time surface deformation data; adjust the coating spray of the coater nozzle according to the real-time surface deformation data and coating compensation data to generate coater nozzle data; use the real-time surface deformation data and coater nozzle data to perform real-time path optimization, so as to generate real-time coating thickness adjustment data; A deviation comparison module is used to train a model with the real-time coating thickness adjustment data to generate a coating thickness prediction model; import the real-time coating thickness adjustment data into the coating thickness prediction model for thickness prediction to generate coating thickness prediction data; calculate the difference between the coating thickness prediction data and the real-time coating thickness prediction data to obtain coating thickness deviation data; compare the coating thickness deviation data with a preset coating thickness deviation threshold to generate a coater control report.

Images