Large-scale rotary equipment blade three-dimensional profile measurement and point cloud splicing method and system based on line laser

Abstract

The invention discloses a large-scale rotary equipment blade three-dimensional profile measurement and point cloud splicing method and system based on line laser, and belongs to the technical field of precision measurement and computer image processing. Reading a CAD model of a to-be-measured blade, extracting geometric features of the blade by a system, and analyzing the geometric features to generate a plurality of measurement path candidate sets; a measured blade is fixed on a rotary table, a multi-dimensional measurement movement mechanism is adopted to carry a line laser sensor to scan the surface of the blade step by step according to a planned final path, and accurate three-dimensional point cloud data is obtained; the method comprises the following steps: after acquiring point cloud data of a blade, initially aligning the scanned point cloud data in combination with position information of equipment; the optimized ICP algorithm is used, a transformation matrix between the point clouds is continuously adjusted through an iteration mode, the distance between the point clouds is minimized, mismatching and noise influences are further reduced, and therefore accurate registration is achieved. The invention aims to solve the problems of low precision and poor efficiency of aero-engine blade three-dimensional profile measurement.

Description

Technical Field

[0001] This invention belongs to the field of precision measurement and computer image processing technology, and specifically relates to a method and system for measuring the three-dimensional profile and stitching point cloud of large rotating equipment blades based on line laser. Background Technology

[0002] In aero-engine manufacturing, blades are critical high-precision components, and the accurate measurement of their surface shape directly affects assembly precision and overall performance. Current blade measurement methods mainly rely on contact measuring tools or single-angle laser scanning, which suffer from low measurement accuracy and poor efficiency.

[0003] In existing technologies, laser scanning path planning methods lack sufficient intelligence and adaptability, resulting in low measurement efficiency and affecting the integrity and accuracy of measurements. Furthermore, when dealing with parts with complex geometries, traditional point cloud stitching algorithms are prone to introducing errors, leading to inaccurate point cloud reconstruction results. Therefore, there is an urgent need for multi-angle, multi-viewpoint line laser scanning path planning methods and high-precision point cloud stitching algorithms to improve measurement accuracy and efficiency. Summary of the Invention

[0004] This invention provides a method and system for measuring the three-dimensional profile and stitching point cloud of large rotating equipment blades based on line laser, in order to solve the problems of low accuracy and poor efficiency in measuring the three-dimensional profile of aero-engine blades.

[0005] This invention is achieved through the following technical solution: A method for measuring the three-dimensional profile and stitching point clouds of blades in large rotating equipment based on line laser, the method comprising the following steps: Step 1: Path planning. The system reads the CAD model of the blade to be tested, extracts the geometric features of the blade, analyzes the geometric features, and generates a set of multiple candidate measurement paths. Step 2: Perform measurements and obtain data. Fix the blade to be measured on the turntable, and use a multi-dimensional measurement motion mechanism equipped with a line laser sensor to scan the surface of the blade step by step according to the planned final path to obtain accurate three-dimensional point cloud data. Step 3: 3D point cloud reconstruction and stitching. After acquiring the point cloud data of the blade, the point cloud data obtained by scanning is initially aligned by combining the equipment's position information. Step 4: Using the optimized ICP algorithm, the transformation matrix between point clouds is continuously adjusted iteratively to minimize the distance between point clouds, further reducing mismatches and noise effects, thereby achieving accurate registration.

[0006] Furthermore, step one specifically involves: each measurement path corresponding to a scanning area to ensure coverage of all detailed areas of the blade; and using a particle swarm optimization algorithm combined with sensor parameters to locally optimize the path to ensure that each measurement path meets the scanning constraints. During the optimization process, the path is corrected based on the measurement error model of the laser scanner to minimize the error:

[0007] in, For scan depth, For the exterior angle of the face, It is an interior angle; After path optimization is completed, the final measurement path will be output to ensure that the measurement efficiency and accuracy of each path are maximized.

[0008] 3. The method according to claim 1, wherein step two specifically comprises: the scanning data includes the X, Y, and Z coordinate information of each measurement point and its corresponding curvature features;

[0009] in, It is the first i One measurement point, , , It is the first i The three-dimensional coordinates of the measurement points It is the curvature feature of that point.

[0010] Furthermore, step three specifically involves: calculating the preliminary transformation matrix of the point cloud at each scanning position using the known device parameters, and applying these transformation matrices to perform geometric transformations on the point cloud data so that the point cloud is roughly aligned. For point It can be translated and rotated using the transformation matrix T:

[0011] in, It is the transformed point; After initial alignment, curvature feature continuity optimization is employed to minimize the difference in normal direction between adjacent point clouds:

[0012] in, , The objective function is the normal to adjacent points. This objective function ensures the continuity of curvature information between adjacent point clouds, thereby avoiding splicing errors caused by local shape changes and ensuring that the point clouds are seamlessly connected in geometric structure after splicing.

[0013] Furthermore, step four specifically involves minimizing the following error function: The objective of the ICP algorithm is to minimize the following error function:

[0014] in, It is a rotation matrix. It is a translation vector. and These are matching points in two point clouds. Represents Euclidean distance; After the above-mentioned refinement process, the resulting point cloud data will have a high-precision three-dimensional structure and can be seamlessly stitched into a complete three-dimensional model of the blade.

[0015] A system for measuring the three-dimensional profile and stitching the point cloud of a large rotating equipment blade based on line laser, characterized in that the system uses the aforementioned method for measuring the three-dimensional profile and stitching the point cloud of a large rotating equipment blade based on line laser, and the system includes: Path planning module: Reads the CAD model of the blade to be tested, extracts the geometric features of the blade, analyzes the geometric features and generates a set of multiple measurement path candidates; Data acquisition module: Performs measurements and acquires data. The blade under test is fixed on the turntable. A multi-dimensional measurement motion mechanism equipped with a line laser sensor scans the surface of the blade step by step according to the planned final path to obtain accurate three-dimensional point cloud data. 3D point cloud reconstruction and stitching module: After acquiring the point cloud data of the blade, the point cloud data obtained by scanning is initially aligned by combining the position information of the equipment. Precise registration module: Using the optimized ICP algorithm, the transformation matrix between point clouds is continuously adjusted iteratively to minimize the distance between point clouds, further reducing mismatches and noise effects, thereby achieving precise registration.

[0016] Furthermore, the path planning module works as follows: each measurement path corresponds to a scanning area to ensure that all detailed areas of the blade are covered; the particle swarm optimization algorithm is used in combination with the sensor parameters to locally optimize the path to ensure that each measurement path can meet the scanning constraints. During the optimization process, the path is corrected based on the measurement error model of the laser scanner to minimize the error:

[0017] in, For scan depth, For the exterior angle of the face, It is an interior angle; After path optimization is completed, the final measurement path will be output to ensure that the measurement efficiency and accuracy of each path are maximized.

[0018] Furthermore, the data acquisition module operates as follows: the scanned data includes the X, Y, and Z coordinate information of each measurement point and its corresponding curvature features:

[0019] in, It is the first i One measurement point, , , It is the first i The three-dimensional coordinates of the measurement points It is the curvature feature of that point.

[0020] Furthermore, the working process of the three-dimensional point cloud reconstruction and stitching module is as follows: using known equipment parameters, the preliminary transformation matrix of the point cloud at each scanning position is calculated, and these transformation matrices are applied to perform geometric transformation on the point cloud data so that the point cloud is roughly aligned. For point It can be translated and rotated using the transformation matrix T:

[0021] in, It is the transformed point; After initial alignment, curvature feature continuity optimization is employed to minimize the difference in normal direction between adjacent point clouds:

[0022] in, , The objective function is the normal to adjacent points. This objective function ensures the continuity of curvature information between adjacent point clouds, thereby avoiding splicing errors caused by local shape changes and ensuring that the point clouds are seamlessly connected in geometric structure after splicing.

[0023] Furthermore, the precise registration module operates as follows: the objective of the ICP algorithm is to minimize the following error function:

[0024] in, It is a rotation matrix. It is a translation vector. and These are matching points in two point clouds. Represents Euclidean distance; After the above-mentioned refinement process, the resulting point cloud data will have a high-precision three-dimensional structure and can be seamlessly stitched into a complete three-dimensional model of the blade.

[0025] The beneficial effects of this invention are: 1. The multi-angle linear laser scanning path planning method proposed in this invention optimizes the scanning path by combining the geometric features of the blade. Unlike traditional scanning path planning methods, this invention can fully cover the complex surface of the blade from multiple scanning perspectives, ensuring the integrity and accuracy of point cloud data. Through particle swarm optimization algorithm, this method can efficiently generate the optimal scanning path and effectively avoid redundant scanning, improving scanning efficiency and reducing measurement time; 2. Compared to existing point cloud stitching methods, this invention achieves higher precision point cloud stitching by combining device information with a curvature continuity optimization algorithm. Curvature feature optimization ensures seamless connection of the point cloud in terms of geometric structure, avoiding stitching errors caused by changes in scanning angle and surface curvature discontinuities. Furthermore, this invention uses an optimized ICP (Iterative Closest Point) algorithm for refined registration of the point cloud, further reducing mismatches and noise effects, and improving stitching accuracy.

[0026] 3. The trajectory planning and point cloud stitching method of this invention can effectively preserve the detailed information of the blade surface and will not lose useful data during the optimization process. This ensures the accuracy of the blade's three-dimensional contour point cloud data and makes the data more suitable for subsequent evaluation, analysis, and visualization. Attached Figure Description

[0027] Figure 1 This is a flowchart of the three-dimensional profile measurement and point cloud stitching process of the blade according to the present invention.

[0028] Figure 2 This is a schematic diagram of the high-precision point cloud model of the three-dimensional profile of the blade of the present invention. Detailed Implementation In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods are omitted so as not to obscure the description of this application with unnecessary detail.

[0029] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0030] Example 1 This embodiment provides a method for measuring the 3D contour and stitching point cloud of a large rotating equipment blade based on line laser scanning. First, a multi-angle, multi-viewpoint line laser scanning path planning method ensures that complete 3D contour data of the complex blade surface can be acquired in multiple scanning directions. Next, the blade's 3D contour is scanned according to the scanning path determined by the path planning to obtain measurement data. Then, a high-precision point cloud stitching algorithm is used to register the scanned point cloud data, thereby obtaining a high-precision 3D contour point cloud model of the blade.

[0031] like Figure 1 As shown, the method includes the following steps: Step 1: Path planning. The system reads the CAD model of the blade to be measured, extracts the geometric features of the blade (surface, boundary, etc.), and analyzes the geometric features to generate a set of multiple measurement path candidates. Step 2: Perform measurements and obtain data. Fix the blade to be measured on the turntable, and use a multi-dimensional measurement motion mechanism equipped with a line laser sensor to scan the surface of the blade step by step according to the planned final path to obtain accurate three-dimensional point cloud data. Step 3: 3D point cloud reconstruction and stitching. After acquiring the point cloud data of the blade, the point cloud data obtained by scanning is initially aligned by combining the equipment's position information (X, Y, Z guide rail positions, turntable angle, sensor pose, etc.). Step 4: Using the optimized ICP (Iterative Closest Point) algorithm, the transformation matrix between point clouds is continuously adjusted iteratively to minimize the distance between point clouds, further reducing mismatches and noise effects, thereby achieving accurate registration.

[0032] Furthermore, step one specifically involves: each measurement path corresponding to a scanning area to ensure coverage of all detailed areas of the blade; to ensure path feasibility, a particle swarm optimization (PSO) algorithm is used in conjunction with sensor parameters to locally optimize the path, ensuring that each measurement path meets the scanning constraints. During the optimization process, the path is corrected based on the measurement error model of the laser scanner to minimize the error:

[0033] in, For scan depth, For the exterior angle of the face, It is an interior angle; After path optimization is completed, the final measurement path will be output to ensure that the measurement efficiency and accuracy of each path are maximized.

[0034] Furthermore, step two specifically involves: the scan data includes the X, Y, and Z coordinate information of each measurement point and its corresponding curvature features.

[0035] in, It is the first i One measurement point, , , It is the first i The three-dimensional coordinates of the measurement points It is the curvature feature of that point.

[0036] Furthermore, step three specifically involves: using known device parameters, the preliminary transformation matrix of the point cloud at each scanning position can be calculated, and these transformation matrices are applied to perform geometric transformations such as translation and rotation on the point cloud data so that the point cloud is roughly aligned; For point It can be translated and rotated using the transformation matrix T:

[0037] in, It is the transformed point.

[0038] Furthermore, after the initial alignment, to further improve the accuracy of point cloud stitching, curvature feature continuity optimization is employed to minimize the difference in normal direction between adjacent point clouds:

[0039] in, , The objective function is the normal to adjacent points. This objective function ensures the continuity of curvature information between adjacent point clouds, thereby avoiding splicing errors caused by local shape changes and ensuring that the point clouds are seamlessly connected in geometric structure after splicing.

[0040] Furthermore, step four specifically involves minimizing the following error function:

[0041] in, It is a rotation matrix. It is a translation vector. and These are matching points in two point clouds. Represents Euclidean distance; After the above-mentioned refinement process, the resulting point cloud data will have a high-precision three-dimensional structure and can be seamlessly stitched into a complete three-dimensional model of the blade.

[0042] Example 2 This embodiment provides a three-dimensional profile measurement and point cloud stitching system for large rotating equipment blades based on line laser. The system uses a three-dimensional profile measurement and point cloud stitching method for large rotating equipment blades based on line laser as described in Embodiment 1. The system includes: Path planning module: Reads the CAD model of the blade to be tested, extracts the geometric features of the blade (surface, boundary, etc.), analyzes the geometric features to generate a set of multiple measurement path candidates; Data acquisition module: Performs measurements and acquires data. The blade under test is fixed on the turntable. A multi-dimensional measurement motion mechanism equipped with a line laser sensor scans the surface of the blade step by step according to the planned final path to obtain accurate three-dimensional point cloud data. 3D point cloud reconstruction and stitching module: After acquiring the point cloud data of the blade, the module first combines the equipment's position information (X, Y, Z guide rail positions, turntable angle, sensor pose, etc.) to perform preliminary alignment of the scanned point cloud data. Precise registration module: Using the optimized ICP (Iterative Closest Point) algorithm, the transformation matrix between point clouds is continuously adjusted iteratively to minimize the distance between point clouds, further reducing mismatches and noise effects, thereby achieving precise registration.

[0043] Furthermore, the path planning module works as follows: each measurement path corresponds to a scanning area to ensure that all detailed areas of the blade are covered; to ensure path feasibility, a particle swarm optimization (PSO) algorithm is used in conjunction with sensor parameters to locally optimize the path, ensuring that each measurement path meets the scanning constraints. During the optimization process, the path is corrected based on the measurement error model of the laser scanner to minimize the error:

[0044] in, For scan depth, For the exterior angle of the face, It is an interior angle; After path optimization is completed, the final measurement path will be output to ensure that the measurement efficiency and accuracy of each path are maximized.

[0045] Furthermore, the data acquisition module operates as follows: the scanned data includes the X, Y, and Z coordinate information of each measurement point and its corresponding curvature features:

[0046] in, It is the first i One measurement point, , , It is the first i The three-dimensional coordinates of the measurement points It is the curvature feature of that point.

[0047] Furthermore, the working process of the three-dimensional point cloud reconstruction and stitching module is as follows: using known equipment parameters, the preliminary transformation matrix of the point cloud at each scanning position can be calculated, and these transformation matrices are applied to perform geometric transformations such as translation and rotation on the point cloud data so that the point cloud is roughly aligned. For point It can be translated and rotated using the transformation matrix T:

[0048] in, It is the transformed point; After initial alignment, to further improve the accuracy of point cloud stitching, curvature feature continuity optimization is employed to minimize the difference in normal direction between adjacent point clouds:

[0049] in, , The objective function is the normal to adjacent points. This objective function ensures the continuity of curvature information between adjacent point clouds, thereby avoiding splicing errors caused by local shape changes and ensuring that the point clouds are seamlessly connected in geometric structure after splicing.

[0050] Furthermore, the precise registration module operates as follows: the objective of the ICP algorithm is to minimize the following error function:

[0051] in, It is a rotation matrix. It is a translation vector. and These are matching points in two point clouds. Represents Euclidean distance; After the above-mentioned refinement process, the resulting point cloud data will have a high-precision three-dimensional structure and can be seamlessly stitched into a complete three-dimensional model of the blade.

Claims

1. A method for measuring the three-dimensional profile and stitching point clouds of blades in large rotating equipment based on line laser, characterized in that, The method includes the following steps: Step 1: Path planning. The system reads the CAD model of the blade to be tested, extracts the geometric features of the blade, analyzes the geometric features, and generates a set of multiple candidate measurement paths. Step 2: Perform measurements and obtain data. Fix the blade to be measured on the turntable, and use a multi-dimensional measurement motion mechanism equipped with a line laser sensor to scan the surface of the blade step by step according to the planned final path to obtain accurate three-dimensional point cloud data. Step 3: 3D point cloud reconstruction and stitching. After acquiring the point cloud data of the blade, the point cloud data obtained by scanning is initially aligned by combining the equipment's position information. Step 4: Using the optimized ICP algorithm, the transformation matrix between point clouds is continuously adjusted iteratively to minimize the distance between point clouds, further reducing mismatches and noise effects, thereby achieving accurate registration.

2. The method according to claim 1, characterized in that, Step one specifically involves: each measurement path corresponds to a scanning area to ensure that all detailed areas of the blade are covered; the particle swarm optimization algorithm is used in combination with the sensor parameters to locally optimize the path to ensure that each measurement path meets the scanning constraints. During the optimization process, the path is corrected based on the measurement error model of the laser scanner to minimize the error: in, For scan depth, For the exterior angle of the face, It is an interior angle; After path optimization is completed, the final measurement path will be output to ensure that the measurement efficiency and accuracy of each path are maximized.

3. The method according to claim 1, characterized in that, Step two specifically involves: scanning data including the X, Y, and Z coordinate information of each measurement point and its corresponding curvature characteristics. in, It is the first i One measurement point, , , It is the first i The three-dimensional coordinates of the measurement points It is the curvature feature of that point.

4. The method according to claim 1, characterized in that, The third step specifically involves: using known device parameters, calculating the preliminary transformation matrix of the point cloud at each scanning position, and applying these transformation matrices to perform geometric transformations on the point cloud data so that the point cloud is roughly aligned; For point It can be translated and rotated using the transformation matrix T: in, It is the transformed point; After initial alignment, curvature feature continuity optimization is employed to minimize the difference in normal direction between adjacent point clouds: in, , The objective function is the normal to adjacent points. This objective function ensures the continuity of curvature information between adjacent point clouds, thereby avoiding splicing errors caused by local shape changes and ensuring that the point clouds are seamlessly connected in geometric structure after splicing.

5. The method according to claim 4, characterized in that, Step four specifically involves minimizing the following error function as the objective of the ICP algorithm: in, It is a rotation matrix. It is a translation vector. and These are matching points in two point clouds. Represents Euclidean distance; After the above-mentioned refinement process, the resulting point cloud data will have a high-precision three-dimensional structure and can be seamlessly stitched into a complete three-dimensional model of the blade.

6. A system for measuring the three-dimensional profile and stitching point clouds of blades in large rotating equipment based on line laser, characterized in that, The system uses a method for measuring the three-dimensional profile and stitching point clouds of large rotating equipment blades based on line lasers as described in any one of claims 1-5. The system includes: Path planning module: Reads the CAD model of the blade to be tested, extracts the geometric features of the blade, analyzes the geometric features and generates a set of multiple measurement path candidates; Data acquisition module: Performs measurements and acquires data. The blade under test is fixed on the turntable. A multi-dimensional measurement motion mechanism equipped with a line laser sensor scans the surface of the blade step by step according to the planned final path to obtain accurate three-dimensional point cloud data. 3D point cloud reconstruction and stitching module: After acquiring the point cloud data of the blade, the point cloud data obtained by scanning is initially aligned by combining the position information of the equipment. Precise registration module: Using the optimized ICP algorithm, the transformation matrix between point clouds is continuously adjusted iteratively to minimize the distance between point clouds, further reducing mismatches and noise effects, thereby achieving precise registration.

7. The system according to claim 6, characterized in that, The path planning module works as follows: each measurement path corresponds to a scanning area to ensure that all detailed areas of the blade are covered; the particle swarm optimization algorithm is used in combination with the sensor parameters to locally optimize the path to ensure that each measurement path meets the scanning constraints. During the optimization process, the path is corrected based on the measurement error model of the laser scanner to minimize the error: in, For scan depth, For the exterior angle of the face, It is an interior angle; After path optimization is completed, the final measurement path will be output to ensure that the measurement efficiency and accuracy of each path are maximized.

8. The method according to claim 6, characterized in that, The data acquisition module works as follows: the scanned data includes the X, Y, and Z coordinate information of each measurement point and its corresponding curvature features: in, It is the first i One measurement point, , , It is the first i The three-dimensional coordinates of the measurement points It is the curvature feature of that point.

9. The method according to claim 6, characterized in that, The working process of the three-dimensional point cloud reconstruction and stitching module is as follows: using known equipment parameters, the initial transformation matrix of the point cloud at each scanning position is calculated, and these transformation matrices are applied to perform geometric transformation on the point cloud data so that the point cloud is roughly aligned. For point It can be translated and rotated using the transformation matrix T: in, It is the transformed point; After initial alignment, curvature feature continuity optimization is employed to minimize the difference in normal direction between adjacent point clouds: in, , The objective function is the normal to adjacent points. This objective function ensures the continuity of curvature information between adjacent point clouds, thereby avoiding splicing errors caused by local shape changes and ensuring that the point clouds are seamlessly connected in geometric structure after splicing.

10. The method according to claim 9, characterized in that, The precise registration module works as follows: The objective of the ICP algorithm is to minimize the following error function: in, It is a rotation matrix. It is a translation vector. and These are matching points in two point clouds. Represents Euclidean distance; After the above-mentioned refinement process, the resulting point cloud data will have a high-precision three-dimensional structure and can be seamlessly stitched into a complete three-dimensional model of the blade.

Images