A method for calculating thickness of thermal barrier coating on turbine blades

Abstract

The present application provides a method for calculating the thickness of a thermal barrier coating of a turbine blade, comprising: optically scanning a turbine blade before and after spraying a thermal barrier coating to obtain a coordinate point cloud set A and a point cloud set B; removing noise points generated during the scanning process of the point cloud set A and the point cloud set B; selecting point clouds at positions where edge plates are not sprayed by a spraying machine from the point cloud set A and the point cloud set B respectively to form point cloud subsets A1 and point cloud subsets B1, and simplifying the number of point clouds in the point cloud subsets A1 and B1; performing coordinate matching of the point cloud sets A and B; establishing point clouds at positions where the thickness of the sprayed thermal barrier coating is measured to form point cloud subsets A2 and point cloud subsets B2, and simplifying the number of point clouds in the point cloud subset B2 to establish a point cloud subset B21; performing coordinate conversion of the point cloud subset B21 to obtain a point cloud subset B3; calculating the distance between each point in the point cloud subset B3 and each point in the point cloud subset A2, and obtaining the minimum distance among the distances, which is the coating thickness at each position point.

Description

Technical Field

[0001] The present application belongs to the field of aero-engine technology, and in particular relates to a method for calculating the thickness of a thermal barrier coating on a turbine blade. Background Art

[0002] In order to reduce the temperature of the turbine blade wall, existing aircraft engines generally adopt the method of spraying thermal barrier coatings with low thermal conductivity on the outer surface of the blade. The thickness of the coating directly affects the temperature of the blade wall, the weight of the blade and the size of the blade shape. Therefore, accurate measurement of the blade coating thickness is of great significance to blade design.

[0003] At present, the non-destructive testing of blade coating thickness mainly includes eddy current method, ultrasonic method, etc. Due to the obstruction between blades, the above methods cannot directly measure the thermal barrier coating on the surface of double or multi-jointed turbine guide blades with complex surfaces. The optical scanning method can achieve the measurement of the obstructed position through multi-angle splicing. However, the existing method of using optical scanning to non-destructively test the coating thickness obtains the coating thickness by measuring the model size before and after spraying. The thickness of only one position point can be obtained each time. In order to obtain the thickness of all spraying positions, the calculation process is cumbersome and inefficient. Summary of the invention

[0004] The purpose of the present application is to provide a method for calculating the thickness of a thermal barrier coating on a turbine blade, so as to solve or alleviate at least one problem in the background technology.

[0005] The technical solution of the present application is: a method for calculating the thickness of a thermal barrier coating of a turbine blade, comprising:

[0006] Optically scan the turbine blades before and after spraying the thermal barrier coating to obtain coordinate point cloud set A and point cloud set B, wherein both point cloud set A and point cloud set B include point clouds at the positions where the thermal barrier coating is sprayed and point clouds at the positions where the thermal barrier coating is not sprayed;

[0007] Remove the noise generated during the scanning process of point cloud set A and point cloud set B;

[0008] Select point clouds at non-spraying machine positions from point cloud set A and point cloud set B to form point cloud subset A1 and point cloud subset B1 respectively, simplify the number of point clouds in point cloud subset A1 and point cloud subset B1 by sampling method, set the number of retained point clouds, repeat twice, the first time select a first number of point clouds that meet the distance requirements to form point cloud subset A11 and point cloud subset B11, the second time select a second number of point clouds that meet the distance requirements to form point cloud subset A12 and point cloud subset B12, wherein the second number is greater than the first number;

[0009] First, the point cloud subsets A11 and B11 are used to perform preliminary iterative registration calculations using 3D model matching technology. When the conditions are met, the point cloud subsets A12 and B12 are used for precise matching to obtain the rotation matrix R and translation matrix T.

[0010] Point clouds of the sprayed thermal barrier coating positions where thickness measurement is required are selected from point cloud sets A and point cloud sets B to form point cloud subsets A2 and point cloud subsets B2, respectively, and the number of point clouds in point cloud subset B2 is reduced to establish point cloud subset B21;

[0011] The coordinates of the point cloud subset B3 are calculated by B3=R•B21+T, so as to realize the coordinate transformation of the point cloud subset B21 in the thickness calculation process;

[0012] The distance between each point in the point cloud subset B3 and each point in the point cloud subset A2 is calculated, and the minimum distance among the distances is obtained, which is the coating thickness of each sprayed thermal barrier coating position point.

[0013] Furthermore, methods for removing noise generated during the scanning process of point cloud set A and point cloud set B include: statistical analysis denoising method, filter method, machine learning or deep learning method, curve fitting method, and clustering algorithm.

[0014] Furthermore, the sampling methods for simplifying the point cloud subset A1 and the point cloud subset B1 include: a unified sampling method, a gridding method, and a voxelization method.

[0015] Furthermore, the 3D model matching technology includes: iterative closest point technology, key point matching method, and feature matching method.

[0016] Furthermore, when the number of point clouds of the point cloud subset B2 is reduced, the number is set according to the point density requirement for measuring the thickness of the thermal barrier coating.

[0017] The calculation method of the thickness of the thermal barrier coating of the turbine blades of the present application can obtain the coating thickness at all positions at one time, greatly improving the calculation efficiency, accurately matching the point cloud models before and after spraying, and improving the calculation accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solution provided by the present application, the following is a brief introduction to the accompanying drawings. Obviously, the accompanying drawings described below are only some embodiments of the present application.

[0019] Figure 1 Schematic diagram of the thermal barrier coating thickness calculation method of this application.

[0020] Figure 2 This is a schematic diagram of a double high-pressure turbine guide vane according to an embodiment of the present application. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solutions and advantages of the implementation of this application clearer, the technical solutions in the embodiments of this application will be described in more detail below in conjunction with the drawings in the embodiments of this application.

[0022] In order to overcome the problem of cumbersome calculation process and low efficiency of the optical scanning method in the prior art, the present application provides a method for calculating the thickness of thermal barrier coating of turbine blades.

[0023] like Figure 1 As shown, the method for calculating the thickness of the thermal barrier coating of a turbine blade provided in this application includes the following process:

[0024] Step S1, optically scanning the turbine blades before and after spraying the thermal barrier coating to obtain a coordinate point cloud set A and a point cloud set B.

[0025] The turbine blade to be measured is placed in a 3D optical scanner, and the measuring light is projected to the thermal barrier spraying position and non-spraying position of the blade coating, and scanned to obtain the spatial dimension information of the blade. The coordinate point cloud before and after spraying is recorded as A and B respectively.

[0026] This application takes the coating thickness of a double high-pressure turbine guide blade of a plasma spray coating as an example. Figure 2 As shown, the actual high-pressure turbine guide blade after spraying the plasma thermal barrier coating is placed in a three-dimensional blue light scanner, and the measuring light is projected onto the edge plate of the thermal barrier coating spraying position 1 and the non-spraying position 2 of the high-pressure turbine guide blade, and scanned to obtain the spatial size information of the high-pressure turbine guide blade and generate a coordinate point cloud set B.

[0027] The thermal barrier coating on the surface of the high-pressure turbine guide blade is removed by chemical removal method and then cleaned. The cleaned high-pressure turbine guide blade is subjected to blue light scanning by the same method to obtain a coordinate point cloud set A.

[0028] Step S2: removing noise points generated during the scanning process of point cloud sets A and B.

[0029] To remove the noise generated during the scanning process of the point cloud set A and B, the methods that can be used include but are not limited to: statistical analysis denoising method, filter method, machine learning or deep learning method, curve fitting method, clustering algorithm, etc.

[0030] For example, in this embodiment of the present application, a method of setting the maximum spacing between point clouds (belonging to the statistical analysis denoising method) is used to remove the noise points generated during the scanning process, and the process includes:

[0031] 1) Load the point cloud data or point cloud set that needs to be denoised;

[0032] 2) For each point in the point cloud, calculate the distance between it and all other points and find the nearest neighbor of each point;

[0033] 3) Calculate the maximum value m among all nearest neighbor distances, take a·m as the threshold, and select the coefficient a in the range of 0.95-0.99;

[0034] 4) For each point, if the distance between it and its nearest neighbor is greater than the threshold, the point is marked as a noise point and removed;

[0035] 5) Repeat the above steps until there is no noise point.

[0036] Step S3, simplifying the point cloud sets A and B to include the point cloud of the non-spraying machine plus position 2.

[0037] In point cloud sets A and B respectively, point clouds at the edge plate of position 2 other than the spraying machine are selected to form point cloud subsets A1 and B1.

[0038] The number of point clouds in point cloud subsets A1 and B1 is reduced by sampling methods, and the number of retained point clouds is set, and the process is repeated twice. The first time, a small number of distant point cloud subsets A11 and B11 are selected; the second time, a larger number of point cloud subsets A12 and B12 are selected. Sampling methods include but are not limited to: unified sampling method, gridding method, voxelization method, etc.

[0039] For example, in this embodiment of the present application, the number of point clouds in the point cloud subsets A1 and B1 is simplified by a weighted random sampling method based on distance uniformity, and the number of retained point clouds is set, which is repeated twice. The first time, a small number of distant point cloud subsets A11 and B11 are selected; the second time, a larger number of point cloud subsets A12 and B12 are selected. The process is:

[0040] 1) Load the point cloud data or point cloud subset that needs to be denoised;

[0041] 2) For each point in the point cloud, calculate the distance between it and all other points and find the nearest neighbor of each point;

[0042] 3) For each point, calculate the average of its distances to all other points, sort them according to the average, and use the minimum-maximum normalization method to convert the average distance of each point into a score;

[0043] 4) Set the sampling rate to 5%, calculate the number of points to be retained, perform weighted random sampling on all points, the weight is the score, the retention strategy is to select points with high scores, and retain the sampled points, which are recorded as A11 and B11 respectively;

[0044] 5) Set the sampling rate to 10%, calculate the number of points to be retained, perform weighted random sampling on all points, the weight is the score, the retention strategy is to select points with high scores, and retain the sampled points, which are recorded as A12 and B12 respectively.

[0045] Step S4: coordinate matching of point cloud sets A and B.

[0046] First, the point cloud subsets A11 and B11 are used to perform preliminary iterative registration calculations using 3D model matching technology. When the conditions are met, point cloud subsets A12 and B12 are used for precise matching to obtain the rotation matrix R and translation matrix T.

[0047] 3D model matching technology includes but is not limited to: iterative closest point (ICP) technology, key point matching method, feature matching method, etc.

[0048] KD-Tree, octree and other algorithms can be used to accelerate the search calculation of the nearest matching point and reduce the complexity of the algorithm.

[0049] For example, in this embodiment of the present application, the point cloud subsets A11 and B11 are implemented by the iterative closest point (ICP) technology to perform preliminary iterations for registration calculation; after the conditions are met, the point cloud subsets A12 and B12 are used for precise matching to obtain the rotation matrix R and the translation matrix T. The matching process is: , When matching, the rotation matrix R and translation matrix T are calculated so that Minimum, , .

[0050] Step S5: Establish thickness calculation point cloud subsets A2 and B21.

[0051] Point clouds at the locations of the sprayed thermal barrier coatings whose thickness needs to be measured are selected from point cloud sets A and B to form point cloud subsets A2 and B2, respectively.

[0052] The number of point clouds in point cloud subset B2 is simplified to establish point cloud subset B21, and the number is set appropriately according to the point density requirement for measuring the thickness of thermal barrier coating. In this embodiment, the sampling ratio is set to 1%, and the simplification method can refer to 1) to 4) in step S3.

[0053] Step S6: perform coordinate transformation of the calculated point cloud subset B21.

[0054] The coordinates of the point cloud subset B3 are calculated by B3=R•B21+T, and the coordinate transformation of the point cloud subset B21 in the thickness calculation process is realized.

[0055] Step S7: Calculate coating thickness.

[0056] Calculate the distance between each point in point cloud subset B3 and each point in point cloud subset A2. Obtain the minimum distance among the distances, which is the coating thickness of each thermal barrier coating position point.

[0057] Table 1 shows the coating thickness at each coordinate point obtained in this embodiment of the present application.

[0058] Table 1 Coating thickness of a certain high guide vane (mm)

[0059] x y z Coating thickness 1 0.49047 3.688373 359.8473 0.41 2 -0.36543 -0.23187 359.5561 0.44 3 -1.2332 -4.14963 359.2655 0.42 4 -2.0325 -7.26548 358.9605 0.40 5 -2.8845 -10.3265 358.6652 0.38 6 -3.6320 -14.8952 358.3624 0.35 7 -4.5024 -17.0245 358.0658 0.33 8 -5.4154 -20.6589 357.7609 0.31 9 -6.3365 -23.1256 357.4657 0.29 10 -7.2202 -26.7568 357.1326 0.26 ...... ...... ...... ...... ......

[0060] The calculation method of the thickness of the thermal barrier coating of the turbine blades of the present application can obtain the coating thickness at all positions at one time, greatly improving the calculation efficiency, accurately matching the point cloud models before and after spraying, and improving the calculation accuracy.

[0061] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A method for calculating the thickness of thermal barrier coating of turbine blades, characterized in that: include: Optically scan the turbine blades before and after spraying the thermal barrier coating to obtain coordinate point cloud set A and point cloud set B, wherein both point cloud set A and point cloud set B include point clouds at the positions where the thermal barrier coating is sprayed and point clouds at the positions where the thermal barrier coating is not sprayed; Remove the noise generated during the scanning process of point cloud set A and point cloud set B; Select point clouds at non-spraying machine positions from point cloud set A and point cloud set B to form point cloud subset A1 and point cloud subset B1 respectively, simplify the number of point clouds in point cloud subset A1 and point cloud subset B1 by sampling method, set the number of retained point clouds, repeat twice, the first time select a first number of point clouds that meet the distance requirements to form point cloud subset A11 and point cloud subset B11, the second time select a second number of point clouds that meet the distance requirements to form point cloud subset A12 and point cloud subset B12, wherein the second number is greater than the first number; First, the point cloud subset A11 and the point cloud subset B11 are used to perform preliminary iterative registration calculations using the 3D model matching technology. When the conditions are met, the point cloud subset A12 and the point cloud subset B12 are used to match to obtain the rotation matrix R and the translation matrix T. Point clouds of the sprayed thermal barrier coating positions where thickness measurement is required are selected from point cloud sets A and point cloud sets B to form point cloud subsets A2 and point cloud subsets B2, respectively, and the number of point clouds in point cloud subset B2 is reduced to establish point cloud subset B21; The coordinates of the point cloud subset B3 are calculated by B3=R•B21+T, so as to realize the coordinate transformation of the point cloud subset B21 in the thickness calculation process; The distance between each point in the point cloud subset B3 and each point in the point cloud subset A2 is calculated, and the minimum distance among the distances is obtained, which is the coating thickness of each sprayed thermal barrier coating position point.

2. The method for calculating the thickness of the thermal barrier coating of a turbine blade according to claim 1, characterized in that: Methods for removing noise generated during the scanning process of point cloud set A and point cloud set B include: statistical analysis denoising method, filter method, machine learning or deep learning method, curve fitting method, and clustering algorithm.

3. The method for calculating the thickness of the thermal barrier coating of a turbine blade according to claim 1, characterized in that: The sampling methods for simplifying the point cloud subset A1 and the point cloud subset B1 include: a unified sampling method, a gridding method, and a voxelization method.

4. The method for calculating the thickness of thermal barrier coating of turbine blades according to claim 1, characterized in that: The 3D model matching technology includes: iterative closest point technology, key point matching method, and feature matching method.

5. The method for calculating the thickness of thermal barrier coating of turbine blades according to claim 1, characterized in that: When the number of point clouds of the point cloud subset B2 is reduced, the number is set according to the point density requirement for measuring the thickness of the thermal barrier coating.

Images