Intelligent metal plate spraying method

Abstract

The invention discloses an intelligent metal plate spraying method, which adopts a mode based on machine vision, three-dimensional modeling, spraying path optimization and real-time quality feedback control, and achieves the purpose of improving spraying precision, uniformity and production efficiency by combining digital modeling, accurate path planning and dynamic feedback adjustment in the spraying process. Therefore, optimization of the automatic and intelligent spraying process is achieved. According to the technical scheme, the problems that in a traditional spraying process, path planning precision is insufficient, and spraying quality cannot be fed back and adjusted in real time are effectively solved, and the consistency of the spraying quality and the automation level of the production process are remarkably improved.

Description

Technical Field

[0001] The invention relates to the field of sheet metal spraying, and in particular to an intelligent sheet metal spraying method. Background Art

[0002] With the development of manufacturing industry, automated spraying technology has been widely used in many industries, especially in sheet metal, automobile painting, home appliance manufacturing, etc. Traditional spraying technology usually relies on manual operation or semi-automatic equipment. Although it can meet production needs in some cases, its unstable spraying quality, low production efficiency, and frequent manual intervention limit the degree of automation of production and the improvement of product quality.

[0003] In the past few years, with the continuous development of intelligent manufacturing technology, automated spraying technology has gradually developed towards higher intelligence and precision. Modern automated spraying systems are increasingly using advanced technologies such as machine vision, sensor technology, robot control, and computer algorithms to achieve precise control and quality monitoring of the spraying process. These technologies can not only improve the uniformity of spraying and coating adhesion, but also reduce paint waste and improve production efficiency.

[0004] However, the existing automated spraying technology still faces some challenges, especially in spraying path planning, quality monitoring and real-time feedback control. Traditional spraying path planning methods are usually simple and difficult to cope with the surface shape of complex workpieces and the changing coating properties; and although the existing quality control and feedback mechanisms can detect some problems, they often cannot achieve real-time, automatic adjustment and optimization, resulting in coating quality that is difficult to meet strict requirements.

[0005] Therefore, how to use intelligent technology to solve the path planning, quality feedback and optimization problems in the spraying process and improve the accuracy, stability and production efficiency of the spraying process has become a key issue that needs to be urgently solved in the current automated spraying technology. Summary of the invention

[0006] The invention discloses an intelligent sheet metal spraying method, which is used to solve relevant technical problems in the background technology.

[0007] The technical solution provided by the present invention is as follows: an intelligent sheet metal spraying method, comprising the following steps:

[0008] S1. Start the machine vision device to obtain the three-dimensional data of the workpiece and upload it to the system. The system generates a digital model of the workpiece based on the three-dimensional data analysis;

[0009] S2, start the machine vision device, obtain the coating characteristics, and upload them to the system;

[0010] S3, the system generates the spraying path, the movement trajectory of the spray gun and the number of spray layers according to the digital model, the coating characteristics and the preset spraying thickness;

[0011] S4, the system simulates the spraying path, the movement trajectory of the spray gun and the number of spray layers to generate the predicted coating characteristic data of the workpiece after each spraying;

[0012] S5, the system issues a command to control the spraying robot to perform a spraying operation and generate a coating;

[0013] S6, obtaining the actual coating feature data in step S3 and uploading it to the system;

[0014] S7, the system compares the actual coating feature data with the corresponding predicted coating data; if the difference between the actual coating feature data and the predicted coating data is within a preset threshold, jump to step S5; otherwise, jump to S6;

[0015] S8. The system regenerates the optimized spraying path, spray gun movement trajectory and number of spray layers according to the compared defects, and jumps to step S4.

[0016] Furthermore, step S1 specifically includes:

[0017] S101, using a laser scanner and stereo vision technology to obtain three-dimensional data of the workpiece, and using point cloud data generated by laser scanning and stereo vision to perform three-dimensional reconstruction of the workpiece surface through multi-camera angle shooting;

[0018] S102, denoising the acquired point cloud data to improve data quality and remove noise caused by environmental factors or sensor errors; the denoising formula is:

[0019] P filtered =GaussianFilter(P raw ,σ)

[0020] Among them, P filtered is the filtered point cloud data, P raw is the original point cloud data, σ is the filtering standard deviation;

[0021] S103, registering multiple groups of point cloud data from different angles to ensure the accuracy and completeness of the three-dimensional model; the formula is:

[0022]

[0023] Among them, R is the rotation matrix and t is the translation vector, which ensures the precise alignment of point cloud data at different angles;

[0024] S104, generating a high-precision three-dimensional digital workpiece model for subsequent spraying path planning; the specific steps are as follows:

[0025] S1041. Given a point cloud P = {p1, p2, ···, p n} and normal N={n1,n2,···,n n};

[0026] S1042. Calculate the surface, Δf = ▽·N at P, where Δf is the Laplace operator of the reconstructed surface and ▽·N is the divergence of the normal field.

[0027] Furthermore, step S2 specifically includes:

[0028] S201. The coating characteristics include the viscosity, particle distribution and rheology of the coating. The physical properties of the coating are collected in real time through a built-in spectrometer and a laser particle size analyzer.

[0029] S202, after the paint characteristic data is standardized, it is uploaded to the system for subsequent analysis and generation of the spraying path.

[0030] Furthermore, step S3 specifically includes:

[0031] S301. Based on the geometric features of the workpiece surface, the shortest path from the starting point to the end point is calculated as the spraying path; the calculation formula is:

[0032]

[0033] Where s and e are the starting and ending points respectively, cost(i) is the cost of each step, and h(i) is the heuristic function;

[0034] S302, using Bezier curve fitting technology to generate the movement trajectory of the spray gun, ensuring that the movement of the spray gun on the surface of the workpiece is smooth and covers the spraying area; the specific formula is:

[0035] B(t)=(1-t) 3 P0+3(1-t) 2 tP1+3(1-t)t 2 P2+t 3 P3

[0036] Among them, P0, P1, P2, P3 are control points, and t is the time parameter;

[0037] S303. According to the rheological properties of the coating and the preset spraying thickness, the system calculates the number of spray layers required for each spraying; the number of layers is calculated using the coating rheological model:

[0038]

[0039] Among them, N layers is the number of spray layers calculated, target thickness is the target coating thickness, and layerthickness per pass is the coating thickness sprayed each time.

[0040] Furthermore, step S4 specifically includes:

[0041] S401. Simulate the coating distribution during the spraying process, consider the spraying pressure, spray gun angle, and spraying distance data, and predict the thickness distribution and uniformity of each layer of coating. The calculation formula is:

[0042] 2p = 0

[0043] Where p represents the spraying pressure field, and the finite element method is used to solve this equation to simulate the spraying process;

[0044] S402, using a fluid dynamics model to simulate the flow behavior of paint particles, predict the adhesion and uniformity of the coating, and generate coating characteristic data after each spraying, the formula is:

[0045]

[0046] Among them, ρ is the density of the coating particles, ν is the fluid velocity field;

[0047] S403. Based on the above simulation results, the system optimizes the model prediction through regression analysis. The formula of the regression model is:

[0048] y=Xβ+∈

[0049] Among them, y is the target variable (such as coating thickness), X is the feature matrix, β is the regression coefficient, and ∈ is the error term.

[0050] Furthermore, step S5 specifically includes:

[0051] S501, the spraying robot is controlled by a real-time operating system to ensure that the spraying robot performs high-precision spraying operations according to the set path;

[0052] S502, the system communicates with the spraying robot to ensure accurate execution of the spraying path and the movement trajectory of the spray gun.

[0053] Furthermore, in step S6, actual coating characteristic data is collected in real time through sensors, including coating thickness, adhesion, and coating uniformity.

[0054] Furthermore, step S7 specifically includes:

[0055] S701. The system uses an error analysis algorithm to calculate the difference between the actual coating characteristics and the predicted coating characteristics. The specific calculation formula is as follows:

[0056]

[0057] Among them, y pred (i) is the predicted coating characteristics, y act (i) is the actual coating characteristics, n is the number of sampling points;

[0058] S702: If the RMSE value is less than the preset threshold, it is considered that the spraying quality meets the requirements, and the process jumps to step S5 to continue the next step of spraying; if the RMSE value is greater than the threshold, the process jumps to step S8 for optimization.

[0059] Further, step S8 specifically includes:

[0060] S801, the system adjusts the spraying path and parameters according to the error between the actual coating and the predicted coating to ensure the spraying accuracy and coating quality;

[0061] S8011. If the coating thickness is unevenly distributed, the system adjusts the travel speed and spraying angle of the spray gun. The specific adjustment strategy is:

[0062]

[0063] Among them, V new is the adjusted spray gun speed, V old is the original spray gun speed, Δthickness is the thickness difference, and threshold is the preset coating uniformity threshold;

[0064] S8022. If the coating adhesion does not meet the standard, the system adjusts the spray pressure or the distance between the spray gun and the workpiece to optimize the adhesion. The specific adjustment strategy is:

[0065]

[0066] Among them, P new is the new spraying pressure, P old is the original spraying pressure, Δadhesion is the adhesion difference, and threshold is the preset adhesion standard;

[0067] S802, adjusting the spraying path, spray gun angle and speed according to the spraying defects, and recalculating the number of spray layers to ensure that the coating quality meets the preset standards after each spraying;

[0068] S8021, path adjustment is done by adjusting the angle of the spray gun and the direction of the spray area. The formula is:

[0069]

[0070] Among them, θ new is the new spray gun angle, θ oldis the original spray gun angle, Δangle is the angle error, and threshold is the acceptable range of the angle error;

[0071] S8022. Spraying speed adjustment can ensure uniform spraying by optimizing the movement path of the spray gun. The specific formula is:

[0072]

[0073] Among them, V adjusted is the spraying speed after adjustment, V initial is the original spraying speed, Δuniformity is the coating uniformity error, and threshold is the preset uniformity standard.

[0074] Compared with the prior art, the present invention has the following beneficial effects:

[0075] (1) The intelligent sheet metal spraying method of the present invention adopts a method based on machine vision, three-dimensional modeling, spraying path optimization and real-time quality feedback control. By combining digital modeling, precise path planning and dynamic feedback adjustment during the spraying process, the purpose of improving spraying accuracy, uniformity and production efficiency is achieved, thereby realizing the optimization of the automated and intelligent spraying process. The technical solution of the present invention effectively solves the problems of insufficient path planning accuracy and the inability to provide real-time feedback and adjustment of spraying quality in the traditional spraying process, and significantly improves the consistency of spraying quality and the automation level of the production process.

[0076] (2) The intelligent sheet metal spraying method of the present invention can ensure the high consistency of the spray gun trajectory and coating distribution during the spraying process through precise three-dimensional modeling and path planning, avoiding the quality problems caused by inaccurate paths or uneven coatings in traditional spraying, thereby significantly improving the spraying quality and accuracy.

[0077] (3) The intelligent sheet metal spraying method of the present invention realizes a high degree of automation of the spraying process through machine vision, automatic path planning and robot control technologies, reducing the reliance on manual operation. The system can automatically adjust the spraying parameters according to real-time data, thereby improving production efficiency, reducing labor costs, and reducing errors and waste in the spraying process.

[0078] (4) The intelligent sheet metal spraying method of the present invention, combined with real-time feedback and intelligent optimization algorithms, can dynamically adjust the spraying path and the movement trajectory of the spray gun during the spraying process to adapt to different workpiece shapes, coating properties and environmental conditions. This adaptive capability greatly enhances the flexibility of the system, enabling it to cope with diverse spraying needs. DETAILED DESCRIPTION

[0079] The present invention is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several changes and improvements can also be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0080] The present invention is an intelligent sheet metal spraying method, comprising the following steps:

[0081] S1. Start the machine vision device, obtain the three-dimensional data of the workpiece, and upload it to the system. The system generates a digital model of the workpiece based on the three-dimensional data analysis; the three-dimensional data of the workpiece is obtained through the machine vision device and a digital model is generated, so that the geometric shape of the workpiece can be accurately captured; the spraying path is planned based on the digital model, avoiding the problem of uneven spraying or insufficient coverage caused by the lack of accurate geometric data in traditional spraying. This step provides the basis for subsequent precise spraying and ensures the accuracy and consistency of spraying. Specifically includes:

[0082] S101. First, start the machine vision equipment, including laser scanners and stereo vision cameras, to obtain the three-dimensional data of the workpiece; these devices generate point cloud data through laser scanning and reconstruct the workpiece surface data through multi-angle shooting; the accuracy of laser scanning can reach micron level, which can accurately capture the complex geometric shapes of the workpiece; the stereo vision camera is used to capture the depth information of the workpiece to supplement the details missing from the laser scanner;

[0083] S102, denoising the acquired point cloud data to improve data quality and remove noise caused by environmental factors or sensor errors; the denoising formula is:

[0084] P filtered =GaussianFilter(P raw ,σ)

[0085] Among them, P filtered is the filtered point cloud data, P raw is the original point cloud data, σ is the filtering standard deviation;

[0086] S103. To ensure the accuracy and integrity of the point cloud data, the iterative closest point (ICP) algorithm is used to align multiple sets of point cloud data from different angles. The ICP algorithm optimizes the alignment results by minimizing the differences between point clouds:

[0087]

[0088] Among them, R is the rotation matrix and t is the translation vector, which ensures the precise alignment of point cloud data at different angles;

[0089] S104. Finally, a high-precision three-dimensional digital workpiece model is generated for subsequent spraying path planning. The specific steps are as follows:

[0090] S1041. Given a point cloud P = {p1, p2, ···, p n} and normal N={n1,n2,···,n n};

[0091] S1042. Calculate the surface, Δf = ▽·N at P, where Δf is the Laplace operator of the reconstructed surface and ▽·N is the divergence of the normal field.

[0092] S2. Start the machine vision device, obtain the paint characteristics, and upload them to the system; it can dynamically adjust the spraying parameters according to the characteristics of different paints to ensure the stability and consistency of the spraying effect; this provides the system with a high degree of adaptability to different paint types and environmental conditions, and further improves the spraying quality. Specifically:

[0093] S201. The coating characteristics include the viscosity, particle distribution and rheology of the coating. The physical properties of the coating are collected in real time through the built-in spectrometer and laser particle size analyzer; the collected coating characteristic data include particle size distribution, surface tension, viscosity, etc., which can provide necessary parameters for subsequent spray path planning and spray gun control.

[0094] S202, after the paint characteristic data is standardized, it is uploaded to the system for subsequent analysis and generation of the spraying path.

[0095] S3. The system generates the spraying path, spray gun movement trajectory and number of spray layers based on the digital model, paint characteristics and preset spraying thickness.

[0096] S301. Based on the geometric features of the workpiece surface, the shortest path from the starting point to the end point is calculated as the spraying path; the calculation formula is:

[0097]

[0098] Where s and e are the starting and ending points respectively, cost(i) is the cost of each step, and h(i) is the heuristic function;

[0099] S302, using Bezier curve fitting technology to generate the movement trajectory of the spray gun, ensuring that the movement of the spray gun on the surface of the workpiece is smooth and covers the spraying area; the specific formula is:

[0100] B(t)=(1-t) 3 P0+3(1-t) 2 tP1+3(1-t)t 2 P2+t 3 P3

[0101] Among them, P0, P1, P2, P3 are control points, and t is the time parameter;

[0102] S303. According to the rheological properties of the coating and the preset spraying thickness, the system calculates the number of spray layers required for each spraying; the number of layers is calculated using the coating rheological model:

[0103]

[0104] Among them, N layers is the number of spray layers calculated, target thickness is the target coating thickness, and layerthickness per pass is the coating thickness sprayed each time.

[0105] S4. The system simulates the spraying path, the movement trajectory of the spray gun and the number of spray layers to generate the predicted coating characteristic data of the workpiece after each spraying. The prediction results can help optimize the spraying process and avoid defects that may occur during the spraying process, such as uneven coating and insufficient thickness. This process ensures the quality of each spray layer, thereby improving the reliability of the entire spraying process. Specifically:

[0106] S401. Simulate the coating distribution during the spraying process, consider the spraying pressure, spray gun angle, and spraying distance data, and predict the thickness distribution and uniformity of each layer of coating. The calculation formula is:

[0107] 2p = 0

[0108] Where p represents the spraying pressure field, and the finite element method is used to solve this equation to simulate the spraying process;

[0109] S402, using a fluid dynamics model to simulate the flow behavior of paint particles, predict the adhesion and uniformity of the coating, and generate coating characteristic data after each spraying, the formula is:

[0110]

[0111] Among them, ρ is the density of the coating particles, ν is the fluid velocity field;

[0112] S403. Based on the simulation results of finite element analysis and fluid dynamics, regression analysis is used to further optimize the prediction of the coating to improve the prediction accuracy. The regression analysis formula is:

[0113] y=Xβ+∈

[0114] Among them, y is the target variable (such as coating thickness), X is the feature matrix, β is the regression coefficient, and ∈ is the error term.

[0115] S5. The system issues a command to control the spraying robot to perform a spraying operation and generate a real coating. The spraying robot is controlled by a real-time operating system (RTOS) to ensure that the spray gun performs spraying accurately according to the set path. This makes the spraying process more efficient and accurate, reducing errors and instability caused by manual operation. Automated control not only improves production efficiency, but also greatly reduces labor costs and enhances the stability of the production line. Specific:

[0116] S501. The spraying robot is controlled by a real-time operating system to ensure that the spraying robot performs high-precision spraying operations according to the set path; the RTOS task scheduling formula is:

[0117] Task(T)=Max(Priority(Task1),Priority(Task2),···)

[0118] Among them, T is the task, and the priority of the task is determined by the RTOS scheduling policy.

[0119] S502, the system communicates with the spraying robot to ensure accurate execution of the spraying path and the movement trajectory of the spray gun.

[0120] S6. Obtain the actual coating characteristic data in step S3 and upload it to the system. Specifically: The characteristic data of the coating is collected in real time by sensors, including coating thickness, adhesion, uniformity, etc. The thickness of the coating is measured by a laser rangefinder or an optical sensor, and the adhesion is evaluated by a tensile test or a shear test.

[0121] S7, the system compares the actual coating feature data with the corresponding predicted coating data; if the difference between the actual coating feature data and the predicted coating data is within the preset threshold, it jumps to step S5; otherwise, it jumps to S6; this closed-loop feedback mechanism can effectively control the spraying quality and ensure that the coating achieves the expected effect; if there is a deviation between the actual coating and the predicted data, the system will automatically adjust the spraying path and the movement trajectory of the spray gun to ensure the consistency and stability of the spraying quality. Specifically:

[0122] S701. Calculate the difference between the actual coating characteristics and the predicted coating characteristics. The specific calculation formula is as follows:

[0123]

[0124] Among them, y pred (i) is the predicted coating characteristics, y act (i) is the actual coating characteristics, n is the number of sampling points;

[0125] The smaller the RMSE value, the smaller the error between the predicted coating and the actual coating, and the better the spraying quality; conversely, the larger the RMSE value, the greater the difference in spraying quality, and there may be problems such as uneven coating and thickness that does not meet the requirements.

[0126] S702: If the RMSE value is less than the preset threshold, it is considered that the spraying quality meets the requirements, and the process jumps to step S5 to continue the next step of spraying; if the RMSE value is greater than the threshold, the process jumps to step S8 for optimization.

[0127] In order to determine whether the spraying quality meets the requirements, it is first necessary to set a preset RMSE threshold, that is, the acceptable error range of the quality; this threshold is usually determined based on factors such as process requirements, coating thickness, coating characteristics, and production requirements. An example is as follows:

[0128] Processes requiring high precision: If the target coating thickness is 100μm and the error tolerance is 5%, the RMSE threshold can be set to 5μm. When the RMSE value is greater than 5μm, the spraying quality is considered to be substandard and it is necessary to proceed to step S8 for optimization;

[0129] Process with normal accuracy requirements: If the target coating thickness is 200μm and the error tolerance is 10%, the RMSE threshold can be set to 20μm.

[0130] Usually, the RMSE threshold is set based on experience, historical data and quality standards, which represents the allowable coating quality deviation range. Setting a reasonable threshold is the key to ensure spraying quality control.

[0131] At this point, the system first determines the reason why the coating does not meet the standards by analyzing the size of the RMSE and related spraying parameters. Common reasons include:

[0132] Inaccurate spray gun trajectory: There may be deviations in the spray gun movement trajectory, resulting in the coating thickness being too thin or too thick in some areas.

[0133] Unreasonable spraying path: The spraying path planning may result in repeated coverage or missed spraying, resulting in uneven coating distribution.

[0134] Mismatched coating properties: Poor coating adhesion or uniformity due to mismatch between the coating’s rheological properties (such as viscosity, particle distribution, etc.) and the spraying parameters.

[0135] The system will analyze and determine the specific defect type based on the RMSE results and relevant sensor feedback information, providing a basis for subsequent optimization.

[0136] S8. The system regenerates the optimized spray path, spray gun movement trajectory and number of spray layers according to the compared defects, and jumps to step S4; when it is found that the spray quality does not meet the standard, the system will optimize the spray path and spray gun movement trajectory according to the comparison results; this automatic optimization mechanism can adjust the spray parameters through real-time data, reduce the probability of defects, and ensure that each layer of spraying meets the quality requirements; the system's adaptive ability enables it to automatically adjust according to different production environments and spraying requirements, thereby improving the spray quality and production efficiency. Specifically:

[0137] S801, the system adjusts the spraying path and parameters according to the error between the actual coating and the predicted coating to ensure the spraying accuracy and coating quality;

[0138] S8011. If the coating thickness is unevenly distributed, the system adjusts the travel speed and spraying angle of the spray gun. The specific adjustment strategy is:

[0139]

[0140] Among them, V new is the adjusted spray gun speed, V old is the original spray gun speed, Δthickness is the thickness difference, and threshold is the preset coating uniformity threshold;

[0141] S8022. If the coating adhesion does not meet the standard, the system adjusts the spray pressure or the distance between the spray gun and the workpiece to optimize the adhesion. The specific adjustment strategy is:

[0142]

[0143] Among them, P new is the new spraying pressure, P old is the original spraying pressure, Δadhesion is the adhesion difference, and threshold is the preset adhesion standard;

[0144] S802, adjusting the spraying path, spray gun angle and speed according to the spraying defects, and recalculating the number of spray layers to ensure that the coating quality meets the preset standards after each spraying;

[0145] S8021, path adjustment is done by adjusting the angle of the spray gun and the direction of the spray area. The formula is:

[0146]

[0147] Among them, θ new is the new spray gun angle, θ old is the original spray gun angle, Δangle is the angle error, and threshold is the acceptable range of the angle error;

[0148] S8022. Spraying speed adjustment can ensure uniform spraying by optimizing the movement path of the spray gun. The specific formula is:

[0149]

[0150] Among them, V adjusted is the spraying speed after adjustment, V initial is the original spraying speed, Δuniformity is the coating uniformity error, and threshold is the preset uniformity standard.

[0151] In the present application, the terms "upper", "lower", "left", "right", "front", "back", "top", "bottom", "inner", "outer", "middle", "vertical", "horizontal", "lateral", "longitudinal" and the like indicate positions or positional relationships based on the positions or positional relationships shown. These terms are mainly used to better describe the present invention and its embodiments, and are not used to limit the indicated devices, elements or components to have a specific orientation, or to be constructed and operated in a specific orientation.

[0152] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. An intelligent sheet metal spraying method, characterized in that: The steps include: S1. Start the machine vision device to obtain the three-dimensional data of the workpiece and upload it to the system. The system generates a digital model of the workpiece based on the three-dimensional data analysis; S2, start the machine vision device, obtain the coating characteristics, and upload them to the system; S3, the system generates the spraying path, the movement trajectory of the spray gun and the number of spray layers according to the digital model, the coating characteristics and the preset spraying thickness; S4, the system simulates the spraying path, the movement trajectory of the spray gun and the number of spray layers to generate the predicted coating characteristic data of the workpiece after each spraying; S5, the system issues a command to control the spraying robot to perform a spraying operation and generate a coating; S6, obtaining the actual coating feature data in step S3 and uploading it to the system; S7, the system compares the actual coating feature data with the corresponding predicted coating data; if the difference between the actual coating feature data and the predicted coating data is within a preset threshold, jump to step S5; Otherwise, jump to S6; S8. The system regenerates the optimized spraying path, spray gun movement trajectory and number of spray layers according to the compared defects, and jumps to step S4.

2. An intelligent sheet metal spraying method as claimed in claim 1, characterized in that: Step S1 specifically includes: S101, using a laser scanner and stereo vision technology to obtain three-dimensional data of the workpiece, and using point cloud data generated by laser scanning and stereo vision to perform three-dimensional reconstruction of the workpiece surface through multi-camera angle shooting; S102, denoising the acquired point cloud data to improve data quality and remove noise caused by environmental factors or sensor errors; the denoising formula is: P filtered =GaussianFilter(P raw ,σ) Among them, P filtered is the filtered point cloud data, P raw is the original point cloud data, σ is the filtering standard deviation; S103, registering multiple groups of point cloud data from different angles to ensure the accuracy and completeness of the three-dimensional model; the formula is: Among them, R is the rotation matrix and t is the translation vector, which ensures the precise alignment of point cloud data at different angles; S104, generating a high-precision three-dimensional digital workpiece model for subsequent spraying path planning; the specific steps are as follows: S1041. Given a point cloud P = {p1, p2, ···, p n } and normal N={n1,n2,···,n n }; S1042, calculation surface, In P, where Δf is the Laplacian operator of the reconstructed surface, is the divergence of the normal field.

3. The intelligent sheet metal spraying method according to claim 1, characterized in that: Step S2 specifically includes: S201. The coating characteristics include the viscosity, particle distribution and rheology of the coating. The physical properties of the coating are collected in real time through a built-in spectrometer and a laser particle size analyzer. S202, after the paint characteristic data is standardized, it is uploaded to the system for subsequent analysis and generation of the spraying path.

4. The intelligent sheet metal spraying method according to claim 1, characterized in that: Step S3 specifically includes: S301. Based on the geometric features of the workpiece surface, the shortest path from the starting point to the end point is calculated as the spraying path; the calculation formula is: Where s and e are the starting and ending points respectively, cost(i) is the cost of each step, and h(i) is the heuristic function; S302, using Bezier curve fitting technology to generate the movement trajectory of the spray gun, ensuring that the movement of the spray gun on the surface of the workpiece is smooth and covers the spraying area; the specific formula is: B(t)=(1-t) 3 P0+3(1-t) 2 tP1+3(1-t)t 2 P2+t 3 P3 Among them, P0, P1, P2, P3 are control points, and t is the time parameter; S303. According to the rheological properties of the coating and the preset spraying thickness, the system calculates the number of spray layers required for each spraying; the number of layers is calculated using the coating rheological model: Among them, N layers is the number of spray layers calculated, target thickness is the target coating thickness, and layerthickness per pass is the coating thickness sprayed each time.

5. The intelligent sheet metal spraying method according to claim 1, characterized in that: Step S4 specifically includes: S401. Simulate the coating distribution during the spraying process, consider the spraying pressure, spray gun angle, and spraying distance data, and predict the thickness distribution and uniformity of each layer of coating. The calculation formula is: Where p represents the spraying pressure field, and the finite element method is used to solve this equation to simulate the spraying process; S402, using a fluid dynamics model to simulate the flow behavior of paint particles, predict the adhesion and uniformity of the coating, and generate coating characteristic data after each spraying, the formula is: Among them, ρ is the density of the coating particles, ν is the fluid velocity field; S403. Based on the above simulation results, the system optimizes the model prediction through regression analysis. The formula of the regression model is: y=Xβ+∈ Among them, y is the target variable (such as coating thickness), X is the feature matrix, β is the regression coefficient, and ∈ is the error term.

6. The intelligent sheet metal spraying method according to claim 1, characterized in that: Step S5 specifically includes: S501, the spraying robot is controlled by a real-time operating system to ensure that the spraying robot performs high-precision spraying operations according to the set path; S502, the system communicates with the spraying robot to ensure accurate execution of the spraying path and the movement trajectory of the spray gun.

7. The intelligent sheet metal spraying method according to claim 1, characterized in that: In step S6, actual coating characteristic data is collected in real time through sensors, including coating thickness, adhesion, and coating uniformity.

8. The intelligent sheet metal spraying method according to claim 1, characterized in that: Step S7 specifically includes: S701. The system uses an error analysis algorithm to calculate the difference between the actual coating characteristics and the predicted coating characteristics. The specific calculation formula is as follows: Among them, y pred (i) is the predicted coating characteristics, y act (i) is the actual coating characteristics, n is the number of sampling points; S702: If the RMSE value is less than the preset threshold, it is considered that the spraying quality meets the requirements, and the process jumps to step S5 to continue the next step of spraying; if the RMSE value is greater than the threshold, the process jumps to step S8 for optimization.

9. The intelligent sheet metal spraying method according to claim 1, characterized in that: Step S8 specifically includes: S801, the system adjusts the spraying path and parameters according to the error between the actual coating and the predicted coating to ensure the spraying accuracy and coating quality; S8011. If the coating thickness is unevenly distributed, the system adjusts the travel speed and spraying angle of the spray gun. The specific adjustment strategy is: Among them, V new is the adjusted spray gun speed, V old is the original spray gun speed, Δthickness is the thickness difference, and threshold is the preset coating uniformity threshold; S8022: If the coating adhesion does not meet the standard, the system adjusts the spray pressure or the distance between the spray gun and the workpiece to optimize the adhesion. The specific adjustment strategy is: Among them, P new is the new spraying pressure, P old is the original spraying pressure, Δadhesion is the adhesion difference, and threshold is the preset adhesion standard; S802, adjusting the spraying path, spray gun angle and speed according to the spraying defects, and recalculating the number of spray layers to ensure that the coating quality meets the preset standards after each spraying; S8021, path adjustment is done by adjusting the angle of the spray gun and the direction of the spray area. The formula is: Among them, θ new is the new spray gun angle, θ old is the original spray gun angle, Δangle is the angle error, and threshold is the acceptable range of the angle error; S8022. Spraying speed adjustment can ensure uniform spraying by optimizing the movement path of the spray gun. The specific formula is: Among them, V adjusted is the spraying speed after adjustment, V initial is the original spraying speed, Δuniformity is the coating uniformity error, and threshold is the preset uniformity standard.

Images