A complex curved surface coating thickness anisotropy three-dimensional reconstruction method and system based on manifold geodesic metric

Abstract

The application discloses a kind of complex curved surface coating thickness anisotropy three-dimensional reconstruction method and system based on manifold geodesic metric, belong to nondestructive testing and digital modeling technical field.The method is first obtained by light heat thickness and three-dimensional scanning alignment discrete data, subsequently constructs the surface of workpiece manifold grid, utilize the geodesic distance in differential geometry to replace traditional Euclidean space straight line distance as the measurement benchmark of space correlation.On this basis, combined with the construction of anisotropic variation function of spraying process characteristics, and realize the accurate reconstruction of full-field thickness by kriging interpolation.The application solves the problem of interpolation distortion and thickness distribution prediction inaccuracy caused by ignoring the topological structure of curved surface when traditional space interpolation algorithm faces large curvature, complex manifold workpieces such as aviation blades, significantly improves the geometric accuracy and physical consistency of complex curved surface coating full-field detection.

Description

I. Technical Field This invention relates to the fields of nondestructive testing and three-dimensional digital modeling, and in particular to a method and system for anisotropic three-dimensional reconstruction of the thickness of complex curved surface coatings based on manifold geodesics. II. Background Technology Thermal barrier coatings (TBCs) are widely used in high-temperature core components such as turbine blades of aero-engines, and the uniformity of their thickness distribution directly determines the service life and thermal protection performance of the equipment. Photothermal radiation (PTR) thickness measurement technology has become an important method for coating inspection due to its advantages such as non-contact and high precision. In recent years, combining photothermal technology with automated 3D scanners (such as the Coatpro thickness measurement system) to acquire discrete point clouds and thickness data and perform full-field 3D imaging has become an industry trend.

[0001] The drawback of existing technologies lies in the fact that after acquiring discrete measurement points, spatial interpolation algorithms (such as Kriging interpolation and inverse distance weighted interpolation) are typically needed to reconstruct the thickness of unmeasured areas. However, existing geostatistical interpolation methods are all based on three-dimensional Euclidean spatial straight-line measurements (i.e., using spatial straight-line distance to measure data correlation). When dealing with complex manifold components with large curvature and sharp angles (such as the leading and trailing edges of blades), Euclidean distance can produce a severe "spatial short-circuiting" phenomenon (i.e., points physically located on opposite sides of the blade are very close in a straight line, but their actual physical topological distances are extremely far). Forcibly introducing Euclidean measurements can lead to incorrect interpolation weight allocation, causing local data transmission and topological distortion. Furthermore, traditional interpolation models are mostly isotropic models, ignoring the anisotropic characteristics of thickness distribution exhibited by processes such as plasma spraying (APS) in specific spray gun sweeping directions. III. Summary of the Invention (a) Technical problems to be solved The purpose of this invention is to overcome the interpolation distortion problem caused by the use of Euclidean distance and isotropic models in the thickness reconstruction of complex curved surface coatings in the prior art, and to provide a three-dimensional anisotropic reconstruction method for the thickness of complex curved surface coatings based on manifold geodesic measurement.

[0002] (II) Technical Solution This invention is achieved through the following steps: S1 Multidimensional data acquisition and alignment; S2 Manifold topology reconstruction; S3 Geodesic distance field calculation; S4 Anisotropic variogram construction; S5 Manifold kriging interpolation; S6 Three-dimensional thickness contour mapping.

[0003] (III) Beneficial Effects Completely eliminate "spatial short-circuit" error: Use manifold geodesic measurement from differential geometry to replace Euclidean measurement, strictly follow the true geometric topology of complex surfaces, and completely eliminate erroneous cross-interference of blade front and back data when processing the equal curvature region of the blade leading edge.

[0004] Fitting Physical Spraying Priors: For the first time, an anisotropic tensor is introduced in the manifold metric space. The model can adaptively match the directional texture patterns left by the spraying process, and the interpolation fitting accuracy is significantly higher than that of the traditional isotropic model.

[0005] High robustness: The underlying data is combined with a robust regression algorithm to remove outliers, and the high-level algorithm introduces manifold topology, realizing end-to-end noise resistance and bias correction from underlying signal analysis to high-level spatial reconstruction. IV. Description of the attached drawings Figure 1 This is a flowchart of a method provided in an embodiment of the present invention.

[0006] Figure 2 This is a schematic diagram comparing the geodesic distance and the Euclidean distance on a complex blade surface in this invention.

[0007] Figure 3 This is a schematic diagram of the fitting curve of the anisotropic variation function of the manifold in this invention.

[0008] Figure 4 This is a system module structure diagram provided for an embodiment of the present invention. V. Detailed Implementation Methods The technical solution of the present invention will now be clearly and completely described in conjunction with the accompanying drawings and specific embodiments. Step S1: Multidimensional Data Acquisition and Alignment This method utilizes a photothermal thickness measurement system (such as Coatpro) with a six-axis robotic arm in collaboration with a high-precision 3D laser scanner. First, the 3D scanner acquires the geometric point cloud of the surface of the workpiece to be measured (such as a turbine blade). Subsequently, the photothermal thickness measurement system acquires the photothermal phase response signals at discrete points according to a predetermined trajectory.

[0009] To eliminate random noise and periodic electromagnetic interference, frequency-phase difference data of discrete points are extracted when calculating the thickness. Robust core fitting algorithms such as Random Sample Consensus (RANSAC) are then used to extract stable linear features, thereby accurately calculating the coating thickness value at each sampling point. The final result is a high-dimensional aligned dataset containing coordinates and thickness. .

[0010] Step S2: Manifold Topology Reconstruction Discrete point clouds lack surface adjacency relationships. Surface meshing algorithms (such as Poisson surface reconstruction or Delaunay triangulation) are used to reconstruct discrete geometric point clouds into continuous triangular mesh manifold models. ,in For the set of vertices, Let be the set of edges. This allows us to construct a graph-theoretic adjacency matrix of the workpiece surface, establishing its true physical spatial topological relationships.

[0011] Step S3: Calculation of geodesic distance field In the reconstructed manifold mesh Above, calculate any two known thickness sampling points. and Geodesic distance between This embodiment preferably uses the Fast Marching Method (FMM) to solve the Eikonal Equation:

[0012] in, This represents the geodesic distance field. By integrating along the grid surface, the shortest path length that precisely fits the curved surface of the workpiece is obtained, and this is used to replace the Euclidean linear distance. This fundamentally prevents topological distortion caused by Euclidean distance penetrating the interior of the blade.

[0013] Step S4: Construction of anisotropic variation function Because thermal spraying has a dominant spraying direction, the spatial autocorrelation decay rate of the coating thickness distribution differs between directions parallel and perpendicular to the spray gun's movement. A spatial variation function is constructed. :

[0014] Based on this, anisotropic tensors are introduced. Perform an affine transformation on the geodesic distance. Define the primary stroke direction (spraying direction) and the secondary stroke direction to ensure effective correlation distance. Through this step, the variogram can accurately characterize the thickness correlation law that varies with direction on complex curved surfaces.

[0015] Step S5: Manifold Kriging Interpolation For any unmeasured point to be estimated on the manifold surface Its thickness estimate Reconstruction is performed using a linear combination of measurement points:

[0016] Based on the principles of unbiasedness and minimizing estimation variance, the Kriging equations are constructed using the anisotropic geodesic variation function established in step S4, and the optimal weighting coefficients for each sampling point are solved. Because the underlying layer uses geodesic metrics, high weights are strictly assigned to points adjacent to each other along the physical path of the surface, resulting in extremely smooth thickness predictions that are consistent with physical reality.

[0017] Step S6: 3D thickness contour mapping The full-field thickness data obtained through interpolation is used as a texture attribute and mapped back to the 3D manifold mesh model. On the corresponding vertices or faces. Combining pseudo-color rendering algorithms (such as marking extremely thin areas in red and extremely thick areas in blue), a full-field 3D coating cloud map with high-fidelity geometric features and high-precision thickness information is output for subsequent non-destructive evaluation and process feedback.

[0018] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented using computer program-related hardware. The aforementioned computer program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

Claims

1. A method for anisotropic three-dimensional reconstruction of coating thickness on complex curved surfaces based on manifold geodesic metrics, characterized in that, Includes the following steps: S1: Multidimensional data acquisition and alignment: Using a photothermal thickness measurement system and a three-dimensional scanning device, a discrete set of measurement points on the surface of a complex workpiece is acquired. The set of points contains three-dimensional spatial coordinate information and the corresponding coating thickness measurement value. S2: Manifold Topology Reconstruction: Based on the three-dimensional spatial coordinate information, a manifold mesh model of the workpiece surface is established through a point cloud reconstruction algorithm, and an adjacency matrix representing the geometric topological relationship of the workpiece is constructed. S3: Geodesic distance field calculation: On the manifold grid model, the geodesic distance between any two sampling points is calculated using a differential geometry algorithm, which serves as a metric for measuring spatial correlation; S4: Construction of anisotropic variability function: Combining the prior direction information of the workpiece spraying process, an anisotropic variability function model is constructed in the geodesic measurement space. The model considers the autocorrelation differences along different directions of the manifold surface. S5: Manifold Kriging Interpolation: Using geodesic distance as the independent variable, the anisotropic variogram function is used to perform weighted estimation of the non-measured area to reconstruct the coating thickness distribution across the entire field; S6: 3D thickness cloud mapping: The reconstructed thickness data is fused with the manifold mesh model to output a full-field image of the coating thickness with geometric realism.

2. The method according to claim 1, characterized in that, In step S1, the thickness measurement value is obtained by performing linear regression on the photothermal radiation frequency-phase difference data using a robust core fitting algorithm, which is a random sampling consensus algorithm.

3. The method according to claim 1, characterized in that, In step S3, the geodesic distance calculation uses any one of the following: Dijkstra's algorithm, fast travel method, or thermal method.

4. The method according to claim 1, characterized in that, In step S4, the anisotropic variation function is modified by introducing anisotropic tensors to correct the geodesic step size, so that it can simulate the directional characteristics of thickness distribution caused by the spray gun path in processes such as plasma spraying.

5. A complex surface coating thickness reconstruction system for implementing the method of any one of claims 1 to 4, characterized in that, include: Photothermal thickness measurement unit is used to acquire thickness data of discrete points in a non-contact manner; A 3D scanning unit is used to acquire the geometric point cloud of the workpiece surface; The central processing unit is used to execute manifold reconstruction, geodesic calculation, and interpolation reconstruction algorithms.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for anisotropic three-dimensional reconstruction of complex surface coating thickness based on manifold geodesic measurement as described in any one of claims 1 to 4.

7. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor is configured to execute a computer program stored in the memory to implement the steps of the method for anisotropic three-dimensional reconstruction of complex surface coating thickness based on manifold geodesic measurement as described in any one of claims 1 to 4.

Images