Air film hole location positioning method

Abstract

The invention belongs to the field of precision detection, and relates to a film hole location positioning method, which comprises the following steps of: 1) acquiring initial hole location information under a machine tool space coordinate system; 2) predicting the optimal search position of the current to-be-detected hole based on the initial hole position information under the machine tool space coordinate system obtained in the step 1); 3) collecting the hole position information of the optimal search position of the current to-be-detected hole obtained in the step 2), and adjusting the optical equipment; and 4) based on the adjusted optical equipment, performing deviation correction on the position of the predicted hole to complete hole location positioning. The air film hole position positioning method is wide in positioning application range, high in deviation resistance and high in positioning precision.

Description

Technical Field

[0001] This invention belongs to the field of precision testing and relates to a method for locating the position of film pores, and more particularly to a method for locating the position of film pores on turbine blades of aero-engines and gas turbines. Background Technology

[0002] Film cooling holes are tiny pores (typically between 0.2 and 1.0 mm in diameter) on turbine blades of aero engines and gas turbines. Their function is to provide cooling protection for the blades in high-temperature environments, preventing them from burning out. Therefore, the machining quality of the film cooling holes (including their location, diameter, angle, and shape) directly determines the cooling effect and ultimately affects the service life of the blades and the safety performance of the engine. It is a key quality indicator in blade manufacturing and maintenance.

[0003] Automated inspection typically employs optical systems to replace manual labor, enabling rapid and accurate measurement of film-forming holes. In automated optical inspection of film-forming holes, traditional positioning methods rely on converting theoretical hole coordinates to the machine tool coordinate system for addressing and measurement. While widely used, this method suffers from inherent deviations between the theoretical model and the actual workpiece due to casting, manufacturing tolerances, and workpiece deformation. This makes the inspection process susceptible to systematic biases and cumulative errors, impacting the final inspection accuracy and efficiency. However, with increasingly stringent engine performance requirements, the accuracy standards for film-forming hole inspection are also rising, with hole positioning accuracy being particularly critical. Any deviation in hole position measurement directly affects the data quality acquired by the optical device, potentially generating invalid or interfering data and ultimately reducing the accuracy of the final inspection results.

[0004] Meanwhile, automated optical inspection of film vents mainly focuses on specific stages such as vent hole processing, permeability testing, geometric parameter measurement, or vent wall quality analysis. However, there is relatively little publicly available literature on patented technologies and methods for achieving high-precision, adaptive real-time vent hole positioning during automated optical inspection of blades with complex curved surfaces. Summary of the Invention

[0005] In order to solve the above-mentioned technical problems in the background art, the present invention provides a method for locating air film pores that has a wide range of applications, strong anti-deviation ability, and high positioning accuracy.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] A method for locating the position of an air film pore, characterized in that the method includes the following steps:

[0008] 1) Obtain the initial hole position information in the machine tool spatial coordinate system;

[0009] 2) Based on the initial hole position information in the machine tool space coordinate system obtained in step 1), predict the optimal search position of the current hole to be detected;

[0010] 3) Collect the hole position information of the current optimal search position of the hole to be detected obtained in step 2), and adjust the optical equipment;

[0011] 4) Based on the adjusted optical equipment, the position of the predicted hole is corrected to complete the hole positioning.

[0012] The specific implementation method of step 1) above is as follows:

[0013] 1.1) Select N reference points on the blade, obtain the three-dimensional point cloud data of the actual workpiece reference points, perform three-dimensional registration between the theoretical model and the actual point cloud data, and calculate the global rigid transformation matrix T_global from the theoretical coordinate system to the machine tool space coordinate system; where N≥3; and the reference points are ordered non-collinear points.

[0014] 1.2) For the holes to be measured on the blade, extract the three-dimensional position information of each hole in its theoretical design model and coordinate system. theory = (Position = [X, Y, Z], Dir = [I, J, K]); where Position = [X, Y, Z] represents the coordinate point of the hole in three-dimensional space, and X, Y, Z represent the coordinate component values ​​of the X, Y, and Z axes, respectively; where Dir = [I, J, K] represents the direction vector of the hole in three-dimensional space, and I, J, K represent the direction component values ​​of the X, Y, and Z axes, respectively.

[0015] 1.3) Using the global rigid transformation matrix T_global from the theoretical coordinate system to the machine tool coordinate system obtained in step 1.1), the three-dimensional position information of each hole in the theoretical coordinate system obtained in step 1.2) is converted into the initial hole position information Hole in the machine tool space coordinate system. machine The initial hole position information Hole machine The expression is:

[0016] Hole machine =Hole theory ·T_global; The · here refers to matrix multiplication.

[0017] In step 1.1) above, the actual workpiece's three-dimensional point cloud data is acquired by laser scanning and / or probe measurement.

[0018] The specific implementation method of step 2) above is as follows:

[0019] Algorithm input:

[0020] 1) Initial hole location set: Blade machine= {H_1, H_2, ..., H_total} (in machine coordinate system);

[0021] 2) Set of measured actual borehole locations: Blade actual ={H`_1, H`_2, ..., H`_n-1}

[0022] 2.1) When detecting the nth hole, obtain the initial hole position information H_n. Based on the spatial position coordinates, find the K nearest reference holes in the set of measured hole position information, where K = 0 to 3 or 0 to 4.

[0023] 2.2) For each reference hole H`_j, where j∈[1,K], calculate the offset between the initial hole position and the actual measured hole position: δ_j=H`_j-H_j, where the offset is the local deformation error of the air film hole position;

[0024] 2.3) A distance-based weighted average method is used for interpolation to estimate the displacement δ_n of the current nth hole. The calculation formula is as follows:

[0025] Displacement: δ_n=Σ(w_j*δ_j) / Σ(w_j);

[0026] Weight: w_j = 1 / (d_j^p);

[0027] d_j: The distance between the nth hole and the jth reference hole in the theoretical model;

[0028] p: Power parameter; the closer the reference aperture, the greater its weight.

[0029] δ_n: Includes both position offset and hole orientation offset;

[0030] 2.4) The predicted optimal search hole position for the current hole to be detected is obtained by combining the offset of the current hole with the interpolated value:

[0031] Hole prediction =Hole machine _n+δ_n.

[0032] The specific implementation method of step 3) above is as follows:

[0033] On a high-precision five-axis device, based on the hole position information to be measured, Hole prediction After controlling a high-precision optical device to move directly above the hole, data on the hole position is collected to obtain high-quality data predicting the hole position. prediction The high-quality data prediction It is point cloud data or image data.

[0034] The specific implementation method of step 4) above is as follows:

[0035] 4.1) Hole region extraction: Using the optimal search hole position Hole obtained in step 2) for the current hole to be detected. prediction Based on the high-quality data of the predicted aperture obtained in step 3), the baseline is used. prediction Data is extracted from a local area of ​​a hole that is much smaller than the full field of view. ROI ;

[0036] 4.2) Data Preprocessing: Filtering and / or denoising are applied to the local region data obtained in step 4.1). ROI Optimize the data, remove interfering data, and improve data quality;

[0037] 4.3) Hole Boundary Detection and Extraction: Utilizing the geometric features of the air film pores, the local region data after optimization in step 4.2) is extracted. ROI In the process, through feature extraction, template recognition, and / or edge detection, the data of the hole boundary portion is found. border The geometric features of the air film pores are circular or elliptical.

[0038] 4.4) Hole position parameter fitting: Based on the hole boundary data Data obtained in step 4.3). border The least squares algorithm is used to fit a cylindrical or conical model, and the actual position of the hole (Hole) is obtained based on the fitted model. actual Complete the positioning of the air film pores.

[0039] Step 4 above also includes:

[0040] 4.5) The actual position of the current hole obtained in step 4.4) is Hole actual Add to the set of measured actual boreholes Blade actual .

[0041] The advantages of this invention are:

[0042] This invention provides a method for locating the position of a film air hole, comprising: 1) obtaining initial hole position information in a machine tool spatial coordinate system; 2) predicting the optimal search position of the hole to be detected based on the initial hole position information obtained in step 1).

[0043] 3) Collect the hole position information of the current optimal search position of the hole to be detected obtained in step 2), and adjust the optical equipment; 4) Based on the adjusted optical equipment, correct the deviation of the predicted hole position to complete the hole position positioning. The air film hole position positioning method provided by this invention is based on the previous hole prediction positioning strategy, constructs a dynamically changing positioning benchmark, overcomes the influence of the deviation between the theoretical model and the actual workpiece on the measurement, realizes rapid and robust positioning of the air film hole sequence, and has good anti-deviation performance; at the same time, it adopts an actual hole position deviation correction mechanism to compensate and optimize the positioning coordinates, significantly reducing systematic deviation, thereby improving the positioning accuracy of a single hole and laying a reliable foundation for subsequent geometric parameter measurement; in addition, this invention adopts a method to improve the accuracy of hole position, thereby improving the accuracy of the collected data and thus improving the detection accuracy; moreover, this invention adopts a hole position adaptive positioning method to reduce the influence of workpiece casting deformation tolerance, which can significantly improve the detection reliability. The positioning application range of this invention is wide and suitable for the detection of all blade types. Attached Figure Description

[0044] Figure 1 This is a schematic diagram of the process for predicting the position of the hole to be tested, as used in this invention.

[0045] Figure 2 This is a schematic diagram of the process for correcting the deviation of the predicted position of the hole to be tested, as used in this invention.

[0046] Figure 3 This is a flowchart of the air film pore positioning method provided by the present invention. Detailed Implementation

[0047] This invention provides a method for locating air film holes. This method is an adaptive positioning method that combines the prediction and positioning strategy of the previous hole with the real-time deviation correction of the actual hole position. By dynamically predicting and correcting the search position of the next hole through the previous measured hole position, the systematic deviation and cumulative error between the theoretical model and the actual workpiece can be effectively compensated, and the detection quality and automation efficiency can be improved simultaneously.

[0048] See Figure 3 The core technical steps of the air film pore location method provided by the present invention are as follows:

[0049] 1. Global initial positioning, i.e., obtaining the initial hole position information in the machine tool space coordinate system;

[0050] 1) Select certain reference points on the blade, and use laser scanning, probe measurement and other methods to obtain the actual workpiece three-dimensional point cloud data. Perform three-dimensional registration between the theoretical model and the actual point cloud data, and calculate the global rigid transformation matrix (T_global) from the theoretical coordinate system to the machine tool space coordinate system.

[0051] 2) For the holes to be measured on the blade, extract the three-dimensional position information of each hole in its theoretical design model and coordinate system. theory =(Position=[X, Y, Z], Dir=[I, J, K]).

[0052] 3) Obtain the initial hole position information Hole in the machine tool space coordinate system by transforming the hole position information in the theoretical coordinate system to the machine tool coordinate system using the transformation matrix (T_global) from the theoretical coordinate system to the machine tool coordinate system. machine .

[0053] Hole machine =Hole theory .T_global

[0054] 2. Predictive positioning based on the previous hole: This method predicts the optimal search position of the current hole to be inspected based on the initial hole position information in the machine tool's spatial coordinate system. Its core idea is to abandon the traditional blind-walking strategy that relies on theoretical model hole position coordinates, and instead adopt a context-aware sequential predictive positioning method. This method does not completely depend on the coordinates of the theoretical model; it combines real-time measured data of inspected holes with prior knowledge of the theoretical model to dynamically and adaptively predict the optimal search position of the current hole to be inspected.

[0055] Algorithm input:

[0056] 1) Initial hole location set: Blade machine = {H_1, H_2, ..., H_total} (in machine coordinate system);

[0057] 2) Set of measured actual borehole locations: Blade actual ={H`_1, H`_2, ..., H`_n-1}

[0058] See Figure 1 The specific method for predicting the optimal search position of the current hole to be detected is as follows:

[0059] 1) When detecting the nth hole, obtain the initial hole position information H_n. Based on the spatial position coordinates, find the K nearest reference holes (neighboring holes) in the set of measured hole position information, where K = 0 to 3 or 0 to 4;

[0060] 2) For each reference hole H`_j (j∈[1,K]), calculate the offset between the initial hole position and the actual measured hole position: δ_j=H`_j-H_j, this offset represents the local deformation error of the air film hole position;

[0061] 3) The displacement δ_n of the current nth hole is estimated by interpolation using a distance-based weighted average method. The calculation formula is as follows:

[0062] Displacement: δ_n=Σ(w_j*δ_j) / Σ(w_j)

[0063] Weight: w_j = 1 / (d_j^p),

[0064] d_j: The distance between the nth hole and the jth reference hole in the theoretical model;

[0065] p: Power parameter (usually 1 or 2). The closer the reference aperture, the greater its weight;

[0066] δ_n: Includes both position offset and hole orientation offset;

[0067] 4) Obtain the predicted optimal search hole position by combining the offset of the current hole with the interpolated values:

[0068] Hole prediction =Hole machine _n+δ_n

[0069] 3. Collect hole position information of the optimal search position of the hole to be detected, and adjust the optical equipment;

[0070] Hole information of the pore location of the air film to be measured. prediction =(Position=[X, Y, Z], Dir=[I, J, K])

[0071] On a high-precision five-axis device, based on the hole position information to be measured, Hole prediction After controlling the high-precision optical equipment to move directly above the hole, data acquisition is performed to measure the hole position. prediction = (point cloud, image, etc.)

[0072] 4. Actual hole position deviation correction: Based on the adjusted optical equipment, the predicted hole position is corrected to complete the hole position positioning. The core idea of ​​deviation correction and hole position positioning is: in the current hole prediction and positioning stage, the predicted hole position (Hole) is obtained. prediction Once the optical device is moved to this position, the actual aperture is within the field of view of the optical instrument, but it is usually not in the exact center. Using high-precision (micrometer-level) image / point cloud processing algorithms, the actual aperture location is quickly determined within a small area near the predicted location by utilizing the geometric features (circular, elliptical) of the air film aperture through aperture feature extraction and fitting.

[0073] Algorithm input: 1) Predicted hole position information for the current hole: Hole prediction 2) Based on high-quality data (images or point clouds) collected from the predicted borehole locations: Data prediction .

[0074] See Figure 2 The specific methods for deviation correction and hole positioning are as follows:

[0075] 4.1) Hole Region Extraction: Based on the predicted hole location, extract a local area of ​​the hole data that is much smaller than the full field of view. ROI This is used to significantly reduce the amount of data without sacrificing accuracy, and quickly improve the efficiency of subsequent algorithm processing.

[0076] 4.2) Data preprocessing: For local data, algorithms such as filtering and denoising are used for optimization to remove interference data and improve data quality;

[0077] 4.3) Hole Boundary Detection and Extraction: Utilizing the geometric features of the air film pores (circular, elliptical), data is extracted from the local pore regions. ROI In the process, algorithms such as feature extraction, template recognition, and edge detection are used to find the data of the hole boundary portion. border ;

[0078] 4.4) Hole position parameter fitting: Based on hole boundary data... border The least squares algorithm is used to fit a cylindrical or conical model, and the actual position of the hole (Hole) is obtained based on the fitted model. actual ;

[0079] 4.5) Set the current actual hole position Hole actual Add to the set of measured actual boreholes Blade actual .

[0080] 5. Detection of air film pore parameters

[0081] High-precision hole position Hole based on the output of the actual hole position deviation correction process actual And the high-quality image / point cloud data Hole acquired prediction The key geometric parameters of each air film hole (e.g., hole diameter, position, plane angle, yaw angle, etc.) are accurately extracted and calculated, and compared with the design tolerances to comprehensively evaluate the processing quality of the air film hole.

Claims

1. A method for locating the position of air film pores, characterized in that: The air film pore location positioning method includes the following steps: 1) Obtain the initial hole position information in the machine tool spatial coordinate system; 2) Based on the initial hole position information in the machine tool space coordinate system obtained in step 1), predict the optimal search position of the current hole to be detected; 3) Collect the hole position information of the current optimal search position of the hole to be detected obtained in step 2), and adjust the optical equipment; 4) Based on the adjusted optical equipment, the position of the predicted hole is corrected to complete the hole positioning.

2. The air film pore positioning method according to claim 1, characterized in that: The specific implementation method of step 1) is as follows: 1.1) Select N reference points on the blade, obtain the three-dimensional point cloud data of the actual workpiece reference points, perform three-dimensional registration between the theoretical model and the actual point cloud data, and calculate the global rigid transformation matrix T_global from the theoretical coordinate system to the machine tool space coordinate system; where N≥3; and the reference points are ordered non-collinear points. 1.2) For the holes to be measured on the blade, extract the three-dimensional position information of each hole in its theoretical design model and coordinate system. theory = (Position = [X, Y, Z], Dir = [I, J, K]); where Position = [X, Y, Z] represents the coordinate point of the hole in three-dimensional space, and X, Y, Z represent the coordinate component values ​​of the X, Y, and Z axes, respectively; where Dir = [I, J, K] represents the direction vector of the hole in three-dimensional space, and I, J, K represent the direction component values ​​of the X, Y, and Z axes, respectively. 1.3) Using the global rigid transformation matrix T_global from the theoretical coordinate system to the machine tool coordinate system obtained in step 1.1), the three-dimensional position information of each hole in the theoretical coordinate system obtained in step 1.2) is converted into the initial hole position information Hole in the machine tool space coordinate system. machine The initial hole position information Hole machine The expression is: Hole machine =Hole theory ·T_global; The · here refers to matrix multiplication.

3. The air film pore positioning method according to claim 2, characterized in that: In step 1.1), the actual workpiece's three-dimensional point cloud data is acquired by laser scanning and / or probe measurement.

4. The air film pore positioning method according to claim 3, characterized in that: The specific implementation method of step 2) is as follows: Algorithm input: 1) Initial hole location set: Blade machine = {H_1, H_2, ..., H_total} (in machine coordinate system); 2) Set of measured actual borehole locations: Blade actual ={H`_1, H`_2, ..., H`_n-1} 2.1) When detecting the nth hole, obtain the initial hole position information H_n. Based on the spatial position coordinates, find the K nearest reference holes in the set of measured hole position information, where K = 0 to 3 or 0 to 4. 2.2) For each reference hole H`_j, where j∈[1,K], calculate the offset between the initial hole position and the actual measured hole position: δ_j=H`_j-H_j, where the offset is the local deformation error of the air film hole position; 2.3) A distance-based weighted average method is used for interpolation to estimate the displacement δ_n of the current nth hole. The calculation formula is as follows: Displacement: δ_n=Σ(w_j*δ_j) / Σ(w_j); Weight: w_j = 1 / (d_j^p); d_j: The distance between the nth hole and the jth reference hole in the theoretical model; p: Power parameter; the closer the reference aperture, the greater its weight. δ_n: Includes both position offset and hole orientation offset; 2.4) Obtain the predicted optimal search hole position for the current hole to be detected by combining the offset of the current hole with the interpolated value: Hole prediction =Hole machine _n+δ_n。 5. The air film pore positioning method according to claim 4, characterized in that: The specific implementation method of step 3) is as follows: On a high-precision five-axis device, based on the hole position information to be measured, Hole prediction After controlling a high-precision optical device to move directly above the hole, data on the hole position is collected to obtain high-quality data predicting the hole position. prediction The high-quality data prediction It is point cloud data or image data.

6. The air film pore positioning method according to claim 5, characterized in that: The specific implementation method of step 4) is as follows: 4.1) Hole region extraction: Using the optimal search hole position Hole obtained in step 2) for the current hole to be detected. prediction Based on the high-quality data of the predicted aperture obtained in step 3), the baseline is used. prediction Data is extracted from a local area of ​​a hole that is much smaller than the full field of view. ROI ; 4.2) Data Preprocessing: Filtering and / or denoising are applied to the local region data obtained in step 4.1). ROI Optimize the data, remove interfering data, and improve data quality; 4.3) Hole Boundary Detection and Extraction: Utilizing the geometric features of the air film pores, the local region data after optimization in step 4.2) is extracted. ROI In the process, through feature extraction, template recognition, and / or edge detection, the data of the hole boundary portion is found. border The geometric features of the air film pores are circular or elliptical. 4.4) Hole position parameter fitting: Based on the hole boundary data Data obtained in step 4.3). border The least squares algorithm is used to fit a cylindrical or conical model, and the actual position of the hole (Hole) is obtained based on the fitted model. actual Complete the positioning of the air film pores.

7. The air film pore positioning method according to claim 6, characterized in that: Step 4) further includes: 4.5) The actual position of the current hole obtained in step 4.4) is Hole actual Add to the set of measured actual boreholes Blade actual .

Images