Artificial intelligence fusion three-dimensional precision coating method for turbine blade coating preparation

Abstract

This invention discloses an AI-integrated 3D precision coating method for turbine blade coating preparation, belonging to the field of aero-engine hot-end component manufacturing technology. The method includes: collecting full geometric feature data of the turbine blade, optimizing it with an AI modeling algorithm to construct a 3D digital model of the white blade with a global positioning reference; analyzing coating requirements, completing the 3D design of the coating type and generating a 3D coating template model, which is then converted into standardized 2D template data after intelligent detection and confirmation; collecting dual-source feature data, intelligently flattening the curved surfaces in different regions to create a high-precision mold with a global positioning reference; intelligently attaching the mold to the blade and fixing it with vacuum adsorption, verifying the fit, calibrating the spray gun, executing the spraying operation, and monitoring and dynamically adjusting parameters in real time. This invention achieves full-process AI integration from 3D modeling, template design, mold making to intelligent spraying, significantly improving the accuracy and consistency of coating preparation.

Description

Technical Field

[0001] This invention belongs to the field of manufacturing technology for hot-end components of aero-engines and gas turbines, specifically relating to an artificial intelligence-integrated three-dimensional precision coating method for turbine blade coating preparation. Background Technology

[0002] Turbine blades are among the most harsh hot-end components in aero-engines and gas turbines, enduring long-term exposure to high temperatures, high pressures, high speeds, and corrosive gases. To extend blade life and increase operating temperature, thermal barrier coatings, anti-oxidation coatings, or anti-corrosion coatings are typically applied to the surface of turbine blades.

[0003] Traditional turbine blade coating preparation methods mainly rely on manual masking, tape application, or simple tooling molds, which have the following prominent problems: Turbine blades have extremely complex geometry, including fine features such as blade body, leading edge, trailing edge, tenon teeth, film cooling holes, and cooling channels. Traditional methods make it difficult to achieve precise zoning and edge control of the coating.

[0004] Coating pattern design relies on two-dimensional drawings and experience, which makes it difficult to adapt to the continuous changes in the curvature of the three-dimensional surface of the blade, and is prone to defects such as uneven coating thickness, edge lifting or inadequate masking.

[0005] The mold is made with low precision and has poor fit with the blade, which can easily lead to paint leakage or missed spraying during the spraying process.

[0006] The lack of real-time monitoring and dynamic adjustment capabilities during the spraying process results in poor coating quality consistency and a high rework rate.

[0007] In recent years, the development of artificial intelligence and 3D digital technology has provided new technical paths for coating preparation. However, existing technologies have not yet formed a complete, high-precision, and systematic solution that covers the entire process from data acquisition, 3D modeling, pattern design, mold making to intelligent spraying. Summary of the Invention

[0008] The present invention aims to provide an artificial intelligence-integrated three-dimensional precision coating method for turbine blade coating preparation, in order to solve the technical problems of inaccurate coating pattern design, low mold fit, and lack of intelligent control in the spraying process in the prior art.

[0009] To achieve the above objectives, the present invention provides the following technical solution: An artificial intelligence-integrated three-dimensional precision coating method for turbine blade coating preparation includes the following steps: S1: Collect full geometric feature data of turbine blades. After optimization by AI modeling algorithm, the curvature feature of turbine blades is partitioned. After partitioning, a high-precision three-dimensional digital model of turbine blade full features with global positioning reference is built. This model forms a three-dimensional digital model of white blade. S2: Analyze the turbine blade coating requirements, complete the coating type three-dimensional design based on the white blade three-dimensional digital model, and generate a three-dimensional coating pattern model. After intelligent detection and user confirmation, the three-dimensional coating pattern model is converted into standardized two-dimensional pattern data. S3: Collect dual-source feature data from the three-dimensional digital model of the white leaf blade and the standardized two-dimensional pattern data. After the dual-source feature data is intelligently flattened by regional curved surface and the accuracy is verified, mold processing drawings are generated. Based on the mold processing drawings, a high-precision mold with global positioning reference is made. S4: The high-precision mold is intelligently attached to the turbine blade and fixed by vacuum adsorption. After verifying that the fit between the high-precision mold and the turbine blade meets the standard, the spray gun is intelligently calibrated. After calibration, the spraying equipment is started to perform the coating spraying operation. The coating spraying operation process is monitored in real time and the spraying parameters are dynamically adjusted.

[0010] Furthermore, in step S1, the full geometric feature data includes the three-dimensional spatial coordinates, curvature distribution, and dimensional tolerances of the blade body, blade tip, leading edge, trailing edge, tenon teeth, air film vents, and cooling channels. The data acquisition accuracy of the curvature gradient of the blade body along the blade height direction reaches 0.001mm. The AI ​​modeling algorithm performs intelligent noise reduction and completion operations on the full geometric feature data.

[0011] Furthermore, in step S1, the AI ​​modeling algorithm divides the turbine blade curvature feature into a leading edge rounded corner area, a blade body gradient area, a blade tip plane area, and a film film perforation periphery area. The global positioning reference is a crosshair structure, and the global positioning reference is set into three groups, which are located at the blade root, the middle of the blade body, and the blade tip, respectively.

[0012] Further, in step S2, the turbine blade coating requirements are extracted and analyzed by the AI ​​requirement identification module, which classifies the coating into projection coatings and envelope coatings. The coating type 3D design is completed by the AI ​​type design module, which matches the projection coating with 3D projection technology and the envelope coating with 3D envelope technology. The 3D coating template model is marked with coating thickness, contour tolerance, and the distance between the 3D coating template model and the key structure of the turbine blade. The key structure of the turbine blade refers to the core structure on the turbine blade that needs to avoid the coating, specifically including film cooling holes, cooling channels, and tenons.

[0013] The projection coating refers to a coating area located on an approximate plane of the turbine blade or in an area with a gentle curvature change (such as the blade tip plane area or the tenon end face area). The coating pattern can be generated by projecting the two-dimensional design outline onto the blade surface along a specified direction. It is mainly suitable for coating types that allow edge contour errors within ±0.1mm and do not require strict normal uniformity of coating thickness.

[0014] The enveloping coating refers to a coating area located on a complex curved surface region of a turbine blade (such as the blade transition area or the leading edge rounded corner area). The coating pattern needs to be based on a three-dimensional digital model of the white blade. Through operations such as equidistant offset, surface extension, and trimming in three-dimensional space, a three-dimensional curved surface layer that completely covers the blade surface is generated. It is mainly suitable for coating types that require edge contour error ≤ ±0.05mm and coating thickness that is strictly uniformly distributed along the surface normal.

[0015] The aforementioned 3D projection technology refers to: based on the 3D digital model of the blade, establishing a virtual projection reference plane in a 3D software environment that is generally parallel to the area to be coated. The normal direction of this plane is taken as the average direction of the normals of all points in the area or the direction of a principal axis of the global coordinate system. A pre-designed 2D coating pattern is projected onto the surface of the target coating area of ​​the turbine blade along this normal direction. The spatial distance between the projection boundary and key structures such as air film vents and cooling channels is automatically detected. When the distance is less than a preset safety threshold (e.g., 0.3mm-0.8mm), adaptive trimming or offset correction is performed, ultimately generating a 3D coating pattern model that can be used for subsequent flattening and mold making. For projection stretching or compression deformation caused by changes in the curvature of the blade surface, an inverse transformation compensation algorithm based on the curvature distribution of the surface is used for correction. Specifically, based on the mapping relationship between the coordinates of the mesh nodes before and after projection, a local deformation tensor field is established, and the position of each node is solved in reverse iteration until the deviation between the projection boundary and the target contour is less than a set threshold, ensuring that the dimensional error of the coating pattern after projection is controlled within ±0.05mm.

[0016] The aforementioned 3D envelope technology refers to the following: using a 3D digital model of the blade as the target base surface, extracting a surface patch of the area to be coated, and offsetting it outwards at equal intervals along the normal direction of each point on the surface by a predetermined coating thickness value (e.g., 0.20mm-0.30mm), generating an envelope surface that is completely parallel to and equidistant from the original blade surface. This envelope surface is then extended, trimmed, smoothed, and its boundaries fitted, and automatically trimmed to avoid key structures such as air film vents, cooling channels, and tenons (preset avoidance spacing is 0.3mm-0.8mm). The resulting local boundaries are then smoothly transitioned to prevent sharp corners from causing stress concentration in the coating, ultimately forming a precise 3D coating template model that covers the blade surface. This technology ensures the normal uniformity of the coating thickness on complex curved surfaces and is suitable for thermal barrier coatings and anti-oxidation coatings where strict requirements for coating thickness consistency and edge contour accuracy are necessary.

[0017] Furthermore, in step S2, the AI ​​detection and verification module performs intelligent detection on the 3D coating template model. The intelligent detection includes intelligent precision detection, intelligent structural avoidance detection, and intelligent tolerance detection. After the detection, an inspection report is automatically generated. The 3D coating template model and the 3D digital model of the white blade are intelligently fused to generate a 3D preview image of the coating effect. The 3D preview image of the coating effect is delivered to the user for confirmation along with the inspection report.

[0018] Further, in step S2, the three-dimensional coating pattern model is converted into standardized two-dimensional pattern data through the AI ​​format conversion module. The standardized two-dimensional pattern data is in vector format and contains global positioning reference marks, curvature parameters, and dimension annotations. The global positioning reference marks are consistent with the global positioning reference structure in step S1. The AI ​​format conversion module completes the data conversion according to the partition flattening reference.

[0019] Furthermore, in step S3, dual-source feature data is collected through the AI ​​data acquisition module, and the AI ​​flattening algorithm is used to perform intelligent flattening of the curved surface in different regions and verify the accuracy. The AI ​​flattening algorithm matches the Squish+Smash combination algorithm for the leading edge rounded corner area, the UnrollSrfUV dynamic flattening tool for the blade gradient area, the UnrollSrf tool for the blade tip plane area, and the micro-feature conformal flattening algorithm for the area surrounding the air film pores.

[0020] Furthermore, in step S3, the mold processing drawing is in the native format of the engraving equipment and CNC machining equipment. The mold processing drawing has a global positioning reference, curvature parameters, and machining tolerances. The mold material is intelligently matched according to the coating type through the AI ​​material matching module. The high-precision mold is a coating-covered mold or a coating-masked mold.

[0021] Furthermore, in step S4, the AI ​​positioning and matching module completes the intelligent bonding of the high-precision mold and the turbine blade. The AI ​​positioning and matching module controls the vacuum adsorption equipment to achieve vacuum adsorption and fixation of the high-precision mold and the turbine blade. The bonding degree is verified by the AI ​​vision detection module. After the bonding degree verification meets the standard, the spraying calibration stage begins.

[0022] Furthermore, in step S4, the AI ​​spraying calibration module completes the intelligent calibration of the spray gun. This AI spraying calibration module uses the three-dimensional positioning system of the spraying workbench to adjust the spraying path, spraying angle, and spraying distance of the spray gun, so that it accurately matches the shape contour of the high-precision mold and the curved surface features of the turbine blade, ensuring that the calibration meets the standards. During the coating spraying operation, the AI ​​dynamic monitoring module monitors the operation status in real time. This AI dynamic monitoring module calls high-speed vision inspection equipment to perform monitoring tasks, and then the AI ​​control module dynamically adjusts the working parameters of the spray gun according to the monitoring results to ensure the accuracy of coating preparation.

[0023] The beneficial effects of this invention are as follows: 1. High-precision modeling of all geometric features: By collecting curvature gradient data with an accuracy of 0.001mm, and combining AI noise reduction and completion, a three-dimensional digital model of the white leaf with a global positioning reference is constructed, providing a precise digital base for subsequent pattern design and mold making. 2. Intelligent classification and user visualization confirmation: By using AI to identify the coating type (projection type / envelope type), the optimal 3D design technology is matched, and a 3D preview image of the coating effect with avoidance detection is generated, so that users can intuitively confirm the final effect before spraying, which greatly reduces design errors; 3. Adaptive Surface Flattening by Region: For regions with different curvature features such as leading edge rounded corners, blade gradients, blade tip planes, and the periphery of air film pores, algorithms such as Squish+Smash, UnrollSrfUV, UnrollSrf, and micro-feature shape-preserving flattening are used to achieve high-fidelity flattening and ensure the consistency between the two-dimensional pattern data and the three-dimensional surface. 4. Unified global positioning benchmark: A consistent crosshair global positioning benchmark (three sets: blade root, blade middle, and blade tip) is set in the blade 3D model, 2D pattern data, and mold processing drawings to achieve accurate data transmission and physical fit throughout the entire process; 5. Intelligent bonding and real-time monitoring: AI vision inspection is used to verify the bonding degree between the mold and the blade, and vacuum adsorption is used for fixation; during the spraying process, high-speed vision is used to monitor in real time and dynamically adjust the spray gun parameters, which significantly improves the accuracy, consistency and pass rate of coating preparation. Attached Figure Description

[0024] Figure 1 This is a flowchart of the artificial intelligence-integrated three-dimensional precision coating method for preparing turbine blade coatings according to the present invention.

[0025] Figure 2 This is a structural diagram of the complete geometric feature data of the turbine blade of the present invention.

[0026] Figure 3 This is a diagram of the AI ​​modeling algorithm and global positioning benchmark architecture of the present invention.

[0027] Figure 4 This is a flowchart of the AI ​​classification design module of the present invention.

[0028] Figure 5 This is a diagram of the regional adaptation architecture of the AI ​​flattening algorithm of this invention. Detailed Implementation

[0029] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the present invention or its application or use. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0030] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of the invention.

[0031] At the same time, it should be understood that, for ease of description, the dimensions of the various parts shown in the accompanying drawings are not drawn according to actual scale.

[0032] Furthermore, for clarity and brevity, descriptions of well-known structures, functions, and configurations may have been omitted. Those skilled in the art will recognize that various changes and modifications can be made to the examples described herein without departing from the spirit and scope of this disclosure.

[0033] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0034] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0035] like Figure 1 As shown, Figure 1 This is a flowchart of the artificial intelligence-integrated three-dimensional precision coating method for turbine blade coating preparation according to the present invention. The artificial intelligence-integrated three-dimensional precision coating method for turbine blade coating preparation includes the following steps: S1: Collect full geometric feature data of turbine blades. After optimization by AI modeling algorithm, the curvature feature of turbine blades is partitioned. After partitioning, a high-precision three-dimensional digital model of turbine blade full features with global positioning reference is built. This model forms a three-dimensional digital model of white blade. S2: Analyze the turbine blade coating requirements, complete the coating type three-dimensional design based on the white blade three-dimensional digital model, and generate a three-dimensional coating pattern model. After intelligent detection and user confirmation, the three-dimensional coating pattern model is converted into standardized two-dimensional pattern data. S3: Collect dual-source feature data from the three-dimensional digital model of the white leaf blade and the standardized two-dimensional pattern data. After the dual-source feature data is intelligently flattened by regional curved surface and the accuracy is verified, mold processing drawings are generated. Based on the mold processing drawings, a high-precision mold with global positioning reference is made. S4: The high-precision mold is intelligently attached to the turbine blade and fixed by vacuum adsorption. After verifying that the fit between the high-precision mold and the turbine blade meets the standard, the spray gun is intelligently calibrated. After calibration, the spraying equipment is started to perform the coating spraying operation. The coating spraying operation process is monitored in real time and the spraying parameters are dynamically adjusted.

[0036] like Figure 2 As shown, Figure 2 This is a structural diagram of the complete geometric feature data of the turbine blade of the present invention. In step S1, the complete geometric feature data includes the three-dimensional spatial coordinates, curvature distribution, and dimensional tolerances of the blade body, blade tip, leading edge, trailing edge, tenon teeth, air film holes, and cooling channels. The data acquisition accuracy of the curvature gradient of the blade body along the blade height direction reaches 0.001mm. The AI ​​modeling algorithm performs intelligent noise reduction and completion operations on the complete geometric feature data.

[0037] like Figure 3 As shown, Figure 3 This is a diagram of the AI ​​modeling algorithm and global positioning reference architecture of the present invention. In step S1, the AI ​​modeling algorithm divides the turbine blade curvature feature into a leading edge rounded corner area, a blade body gradient area, a blade tip plane area, and a film film aperture periphery area. The global positioning reference is a crosshair structure, and the global positioning reference is set into three groups, which are located at the blade root, the middle of the blade body, and the blade tip, respectively.

[0038] like Figure 4 As shown, Figure 4 This is a flowchart of the AI ​​parting design module of the present invention. In step S2, the turbine blade coating requirements are extracted and analyzed by the AI ​​requirement identification module, which divides the coating into projection coating and envelope coating. The AI ​​parting design module completes the three-dimensional design of the coating parting. The AI ​​parting design module matches the projection coating with three-dimensional projection technology and the envelope coating with three-dimensional envelope technology. The three-dimensional coating template model is marked with the coating thickness, contour tolerance, and the distance between the three-dimensional coating template model and the key structure of the turbine blade. The key structure of the turbine blade refers to the core structure on the turbine blade that needs to avoid the coating, specifically including air film holes, cooling channels, and tenons.

[0039] Furthermore, in step S2, the AI ​​detection and verification module performs intelligent detection on the 3D coating template model. The intelligent detection includes intelligent precision detection, intelligent structural avoidance detection, and intelligent tolerance detection. After the detection, an inspection report is automatically generated. The 3D coating template model and the 3D digital model of the white blade are intelligently fused to generate a 3D preview image of the coating effect. The 3D preview image of the coating effect is delivered to the user for confirmation along with the inspection report.

[0040] Further, in step S2, the three-dimensional coating pattern model is converted into standardized two-dimensional pattern data through the AI ​​format conversion module. The standardized two-dimensional pattern data is in vector format and contains global positioning reference marks, curvature parameters, and dimension annotations. The global positioning reference marks are consistent with the global positioning reference structure in step S1. The AI ​​format conversion module completes the data conversion according to the partition flattening reference.

[0041] like Figure 5 As shown, Figure 5 This is a diagram of the regional adaptation architecture of the AI ​​flattening algorithm of the present invention. In step S3, dual-source feature data is collected through the AI ​​data acquisition module, and regional surface intelligent flattening and accuracy verification are performed through the AI ​​flattening algorithm. The AI ​​flattening algorithm matches the Squish+Smash combination algorithm for the leading edge rounded corner area, the UnrollSrfUV dynamic flattening tool for the blade gradient area, the UnrollSrf tool for the blade tip plane area, and the micro-feature conformal flattening algorithm for the area surrounding the air film pores.

[0042] Furthermore, in step S3, the mold processing drawing is in the native format of the engraving equipment and CNC machining equipment. The mold processing drawing has a global positioning reference, curvature parameters, and machining tolerances. The mold material is intelligently matched according to the coating type through the AI ​​material matching module. The high-precision mold is a coating-covered mold or a coating-masked mold.

[0043] Furthermore, in step S4, the AI ​​positioning and matching module completes the intelligent bonding of the high-precision mold and the turbine blade. The AI ​​positioning and matching module controls the vacuum adsorption equipment to achieve vacuum adsorption and fixation of the high-precision mold and the turbine blade. The bonding degree is verified by the AI ​​vision detection module. After the bonding degree verification meets the standard, the spraying calibration stage begins.

[0044] Furthermore, in step S4, the AI ​​spraying calibration module completes the intelligent calibration of the spray gun. This AI spraying calibration module uses the three-dimensional positioning system of the spraying workbench to adjust the spraying path, spraying angle, and spraying distance of the spray gun, so that it accurately matches the shape contour of the high-precision mold and the curved surface features of the turbine blade, ensuring that the calibration meets the standards. During the coating spraying operation, the AI ​​dynamic monitoring module monitors the operation status in real time. This AI dynamic monitoring module calls high-speed vision inspection equipment to perform monitoring tasks, and then the AI ​​control module dynamically adjusts the working parameters of the spray gun according to the monitoring results to ensure the accuracy of coating preparation.

[0045] This embodiment provides an artificial intelligence-integrated three-dimensional precision coating method for turbine blade coating preparation, the specific steps of which are as follows: Step S1: Construction of 3D Digital Model of White Leaf Leaf A high-precision 3D structured light scanner was used to acquire full geometric feature data of a high-pressure turbine blade for a certain type of aero-engine, including the 3D spatial coordinates, curvature distribution, and dimensional tolerances of the blade body, tip, leading edge, trailing edge, tenon teeth, film cooling holes, and cooling channels. Curvature data was acquired at 0.1mm intervals along the blade height direction, with an acquisition accuracy of 0.001mm.

[0046] The raw point cloud data is input into an AI modeling algorithm, which uses a deep convolutional neural network to intelligently reduce noise (remove outliers and stray reflections) and complete the data (repair missing areas caused by occlusion or reflection). The AI ​​modeling algorithm automatically divides the leaf curvature features into the leading edge rounded corner area, the leaf blade gradient area, the leaf tip plane area, and the area around the air film pores.

[0047] Cross-shaped structures were generated at the blade root, the middle of the blade, and the blade tip as global positioning references. Finally, a high-precision three-dimensional digital model of the turbine blade with all features of these three references was built, namely the three-dimensional digital model of the white blade.

[0048] Step S2: 3D Coating Pattern Design and User Confirmation The 3D digital model of the white blade is imported into the AI ​​requirement recognition module. This module extracts the coating requirements from the process document through semantic parsing: In this embodiment, a thermal barrier coating with a thickness of 0.25 mm is required to be prepared on the blade body and leading edge, while the air film pores, cooling channels, and tenons are left uncoated. The AI ​​requirement recognition module determines this to be an envelope-type coating.

[0049] The AI ​​parting design module employs 3D envelope technology. Using the 3D digital model of the white blade as the base surface, it generates an envelope surface by offsetting outwards by 0.25mm at equal intervals, and automatically trims away key structures such as air film vents, cooling channels, and tenons (with a spacing set to 0.5mm). The generated 3D coating template model is marked with a coating thickness of 0.25±0.02mm and a contour tolerance of ±0.05mm.

[0050] The AI ​​inspection and verification module sequentially performs intelligent precision inspection (checking surface continuity), intelligent structural avoidance inspection (verifying whether the spacing with the air film pores meets the standard), and intelligent tolerance inspection (thickness and contour tolerance) on the 3D coating template model, generating an inspection report. Simultaneously, the 3D coating template model is merged and rendered with the 3D digital model of the white blade to generate a 3D preview image of the coating effect, which is delivered to the user for confirmation along with the inspection report.

[0051] After user confirmation, the AI ​​format conversion module converts the 3D coating pattern model into standardized 2D pattern data (DXF vector format) with global positioning reference marks, curvature parameters, and dimension annotations, based on the partitioned flattening reference of the leading edge rounded corner area, blade gradient area, blade tip plane area, and air film hole periphery area.

[0052] Step S3: High-precision mold making The AI ​​data acquisition module simultaneously extracts dual-source feature data (3D surface information and 2D contour information) from the 3D digital model and standardized 2D template data of the white leaf blade. The AI ​​flattening algorithm performs region-based matching: the leading edge rounded corner area uses a Squish+Smash combined algorithm (first Squish for coarse flattening, then Smash for local relaxation); the blade gradient area uses the UnrollSrfUV dynamic flattening tool (maintaining the curvature gradient in the UV direction); the tip plane area uses the UnrollSrf tool for direct flattening; and the area surrounding the air film pores uses a micro-feature conformal flattening algorithm (locally refining the mesh to ensure the shape of the pore edges is not distorted). After flattening, the area error and boundary alignment are automatically checked, and the checks pass.

[0053] Based on the flattened data, mold processing drawings are generated in both the native format of the engraving machine (G-code) and the native format of the CNC machining center (STEP). The AI ​​material matching module matches high-temperature resistant silicone as the mold material according to the coating type (thermal barrier coating, high spraying temperature). A high-precision coating-masked mold with global positioning reference is produced using a combination of CNC engraving and laser cutting processes.

[0054] Step S4: Intelligent bonding, calibration and spraying operation A high-precision mold is intelligently bonded to the turbine blade using an AI positioning and matching module: the three sets of global positioning references on the blade are aligned with the corresponding reference marks on the mold. The AI ​​positioning and matching module controls a vacuum adsorption device to vacuum-adsorb and fix the mold to the blade surface at an adsorption pressure of -0.08MPa. The AI ​​vision inspection module scans the bonding interface using a binocular vision system and calculates the maximum gap (0.03mm in this embodiment, less than the set threshold of 0.05mm), ensuring the bonding degree meets the standard.

[0055] Entering the spraying calibration stage: The AI ​​spraying calibration module uses the 3D laser positioning system on the spraying workbench to automatically guide the spray gun at the end of the six-axis robotic arm, ensuring that its spraying path, spraying angle (keeping it perpendicular to the mold contour surface within ±2°), and spraying distance (150±5mm) are precisely matched with the high-precision mold profile and the curved surface features of the turbine blades. After calibration, the spraying equipment is started.

[0056] During the coating spraying process, the AI ​​dynamic monitoring module utilizes high-speed visual inspection equipment (sampling frequency 200Hz) to monitor the wet film thickness, edge leakage, and mold adhesion status of the coating in real time. Based on the monitoring results, the AI ​​control module dynamically adjusts the spray gun movement speed (range ±10%) and paint flow rate (range ±5%) to ensure uniform coating thickness and prevent overflow beyond the mold edges. After spraying, the mold is removed, resulting in a turbine blade coating that meets design requirements.

[0057] Comparative Example The same thermal barrier coating was applied to turbine blades of the same model using a traditional manual tape masking method. Results showed that the traditional method required approximately 45 minutes of manual tape application / trimming per blade, resulting in uneven coating edges, a 12% coating intrusion rate around the film cooling holes, and a blade pass rate of only 78%. In contrast, the method described in Example 1 of this invention reduced mold installation time to approximately 3 minutes, produced neat coating edges, eliminated intrusion around the film cooling holes, achieved a first-pass pass rate of 98.5%, and reduced the total processing time per piece by more than 60%.

[0058] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Based on the technical essence of the present invention, any simple modifications, equivalent substitutions, and improvements made to the above embodiments within the spirit and principles of the present invention shall still fall within the protection scope of the present invention.

Claims

1. An artificial intelligence-integrated three-dimensional precision coating method for turbine blade coating preparation, characterized in that, Includes the following steps: S1: Collect full geometric feature data of turbine blades. After optimization by AI modeling algorithm, the curvature feature of turbine blades is partitioned. After partitioning, a high-precision three-dimensional digital model of turbine blade full features with global positioning reference is built. This model forms a three-dimensional digital model of white blade. S2: Analyze the turbine blade coating requirements, complete the coating type three-dimensional design based on the white blade three-dimensional digital model, and generate a three-dimensional coating pattern model. After intelligent detection and user confirmation, the three-dimensional coating pattern model is converted into standardized two-dimensional pattern data. S3: Collect dual-source feature data from the three-dimensional digital model of the white leaf blade and the standardized two-dimensional pattern data. After the dual-source feature data is intelligently flattened by regional curved surface and the accuracy is verified, mold processing drawings are generated. Based on the mold processing drawings, a high-precision mold with global positioning reference is made. S4: The high-precision mold is intelligently attached to the turbine blade and fixed by vacuum adsorption. After verifying that the fit between the high-precision mold and the turbine blade meets the standard, the spray gun is intelligently calibrated. After calibration, the spraying equipment is started to perform the coating spraying operation. The coating spraying operation process is monitored in real time and the spraying parameters are dynamically adjusted.

2. The artificial intelligence-integrated three-dimensional precision coating method for turbine blade coating preparation according to claim 1, characterized in that, In step S1, the full geometric feature data includes the three-dimensional spatial coordinates, curvature distribution, and dimensional tolerances of the blade body, blade tip, leading edge, trailing edge, tenon teeth, air film vents, and cooling channels. The data acquisition accuracy of the curvature gradient of the blade body along the blade height direction reaches 0.001mm. The AI ​​modeling algorithm performs intelligent noise reduction and completion operations on the full geometric feature data.

3. The artificial intelligence-integrated three-dimensional precision coating method for turbine blade coating preparation according to claim 1, characterized in that, In step S1, the AI ​​modeling algorithm divides the turbine blade curvature feature into a leading edge rounded corner area, a blade body gradient area, a blade tip plane area, and a film film hole periphery area. The global positioning reference is a crosshair structure, and the global positioning reference is set into three groups, which are located at the blade root, the middle of the blade body, and the blade tip, respectively.

4. The artificial intelligence-integrated three-dimensional precision coating method for turbine blade coating preparation according to claim 1, characterized in that, In step S2, the turbine blade coating requirements are extracted and analyzed by the AI ​​requirement identification module, which classifies the coating into projection coatings and envelope coatings. The coating type 3D design is completed by the AI ​​type design module, which matches the projection coating with 3D projection technology and the envelope coating with 3D envelope technology. The 3D coating template model is marked with coating thickness, contour tolerance, and the distance between the 3D coating template model and the key structure of the turbine blade. The key structure of the turbine blade refers to the core structure on the turbine blade that needs to avoid the coating, specifically including film cooling holes, cooling channels, and tenons.

5. The artificial intelligence-integrated three-dimensional precision coating method for turbine blade coating preparation according to claim 1, characterized in that, In step S2, the AI ​​detection and verification module performs intelligent detection on the 3D coating template model. The intelligent detection includes intelligent precision detection, intelligent structural avoidance detection, and intelligent tolerance detection. After the detection, an inspection report is automatically generated. The 3D coating template model and the 3D digital model of the white blade are intelligently integrated to generate a 3D preview image of the coating effect. The 3D preview image of the coating effect is delivered to the user for confirmation along with the inspection report.

6. The artificial intelligence-integrated three-dimensional precision coating method for turbine blade coating preparation according to claim 1, characterized in that, In step S2, the three-dimensional coating pattern model is converted into standardized two-dimensional pattern data through the AI ​​format conversion module. The standardized two-dimensional pattern data is in vector format and contains global positioning reference marks, curvature parameters, and dimension annotations. The global positioning reference marks are consistent with the global positioning reference structure in step S1.

7. The artificial intelligence-integrated three-dimensional precision coating method for turbine blade coating preparation according to claim 3, characterized in that, In step S3, dual-source feature data is collected through the AI ​​data acquisition module. The AI ​​flattening algorithm is used to perform intelligent flattening of the curved surface in different regions and verify the accuracy. The AI ​​flattening algorithm matches the Squish+Smash combination algorithm for the leading edge rounded corner area, the UnrollSrfUV dynamic flattening tool for the blade gradient area, the UnrollSrf tool for the blade tip plane area, and the micro-feature conformal flattening algorithm for the area surrounding the air film pores.

8. The artificial intelligence-integrated three-dimensional precision coating method for turbine blade coating preparation according to claim 1, characterized in that, In step S3, the mold processing drawing is in the native format of the engraving equipment and CNC machining equipment. The mold processing drawing has a global positioning reference, curvature parameters, and machining tolerances. The AI ​​material matching module intelligently matches the mold material according to the coating type. The high-precision mold is a coating-covered mold or a coating-masked mold.

9. The artificial intelligence-integrated three-dimensional precision coating method for turbine blade coating preparation according to claim 1, characterized in that, In step S4, the AI ​​positioning and matching module completes the intelligent bonding of the high-precision mold and the turbine blade. The AI ​​positioning and matching module controls the vacuum adsorption equipment to achieve vacuum adsorption and fixation of the high-precision mold and the turbine blade. The bonding degree is verified by the AI ​​vision inspection module. After the bonding degree verification meets the standard, the spraying calibration stage begins.

10. The artificial intelligence-integrated three-dimensional precision coating method for turbine blade coating preparation according to claim 1, characterized in that, In step S4, the AI ​​spraying calibration module completes the intelligent calibration of the spray gun. This AI spraying calibration module uses the three-dimensional positioning system of the spraying workbench to adjust the spraying path, spraying angle, and spraying distance of the spray gun, so that it accurately matches the shape contour of the high-precision mold and the curved surface features of the turbine blade, ensuring that the calibration meets the standards. During the coating spraying operation, the AI ​​dynamic monitoring module monitors the operation status in real time. This AI dynamic monitoring module calls high-speed vision inspection equipment to perform monitoring tasks, and then the AI ​​control module dynamically adjusts the working parameters of the spray gun according to the monitoring results to ensure the accuracy of coating preparation.

Images