Automatic detection and laser additive repairing method for surface defects of metal parts

Abstract

The invention belongs to the technical field of laser additive repair, and discloses a metal part surface defect automatic detection and laser additive repair method which comprises the following steps: (1) scanning the surface of a metal part by adopting a scanner, and preprocessing the obtained point cloud; (2) automatically detecting the defects of sinking, peeling and chipping on the surface of the metal part by analyzing the normal vector of each point in the point cloud, and automatically detecting the crack defect on the surface of the metal part by analyzing the density of the point cloud; (3) carrying out layering treatment on the point cloud corresponding to the sunken or stripped chipping defect, generating a laser scanning path in a sunken or stripped chipping defect area, and carrying out laser cleaning and laser additive repairing on the defect; and (4) according to the point cloud corresponding to the crack defect, a laser scanning path is generated in the crack defect area, and laser cleaning and laser additive repairing are conducted on the crack defect. According to the invention, the surface defects of the metal part can be efficiently detected and repaired, and the repair quality requirement is met.

Description

Technical Field

[0001] The present invention relates to the technical field related to laser additive repair, and in particular to a method for automatic detection of surface defects of metal parts and laser additive repair. Background Art

[0002] Defects such as dents, flaking, and cracks are widely present on the surface of metal parts such as railway rails and automobile brake discs, which seriously affect the service performance of metal parts. Conventional methods for repairing surface defects of metal parts mainly include surfacing and electroplating. However, these methods have disadvantages such as low repair efficiency and poor quality. Laser additive repair technology has the advantages of flexibility and efficiency, high repair accuracy, and good metallurgical bonding of the repair layer. It can be directly used for high-quality repair of defects such as dents, flaking, and cracks on the surface of metal parts. In the process of repairing surface defects of metal parts using laser additive repair technology, it is necessary to locate, classify, manually measure, and model the defects in advance. The whole process is cumbersome and inefficient. In addition, rust, oil stains, etc. often exist in the defect area, which can easily reduce the quality of the laser additive repair layer and affect the repair effect. Therefore, it is urgent to propose a method that can automatically detect surface defects of metal parts, clean rust, oil stains, etc. in the defect area, and realize a high-quality laser additive repair process. Summary of the invention

[0003] The purpose of the present invention is to solve the above-mentioned problems. A method for automatic detection and laser additive repair of surface defects of metal parts is proposed. Automatic detection of dents, peeling and crack defects on the surface of metal parts is achieved by analyzing the surface point cloud of metal parts obtained by scanning with a scanner. The point cloud corresponding to the defect is processed, a laser scanning path is generated in the defect area, the defect is laser cleaned, rust, oil stains, etc. in the defect area are removed, and the defects are laser additive repaired using the generated laser scanning path, thereby achieving efficient and high-quality repair of surface defects of metal parts.

[0004] The present invention is achieved through the following technical solutions:

[0005] A method for automatic detection and laser additive repair of surface defects of metal parts, using a scanner to obtain a point cloud of the metal part surface, automatically detecting depressions, peeling blocks and cracks on the metal part surface by analyzing the point cloud, generating a laser scanning path in the defect area, laser cleaning and laser additive repair of the defects, characterized in that the method comprises the following steps:

[0006] (1) Scan the surface of metal parts with a scanner and pre-process the obtained point cloud;

[0007] (2) Automatically detect dents and peeling defects on the surface of metal parts by analyzing the normal vectors of each point in the point cloud, and automatically detect crack defects on the surface of metal parts by analyzing the point cloud density;

[0008] (3) Performing layered processing on the point cloud corresponding to the dent or peeling defect, generating a laser scanning path in the dent or peeling defect area, and performing laser cleaning and laser additive repair on the defect;

[0009] (4) Based on the point cloud corresponding to the crack defect, a laser scanning path is generated in the crack defect area, and the crack defect is laser cleaned and laser additively repaired.

[0010] Furthermore, in step (2), the normal vector of each point in the surface point cloud of the metal part is calculated based on the K-Nearest Neighbor (KNN) algorithm; the normal vector P of any point P in the point cloud is defined as P The average of the absolute values ​​of the differences between the normal vectors of other points in its neighborhood for:

[0011]

[0012] like If it is greater than the set threshold, then point P is determined to be a dent or peeling defect; traverse all points in the point cloud, filter out all points that belong to dents or peeling defects, complete automatic detection of dents or peeling defects, and obtain the point cloud corresponding to the dent or peeling defect;

[0013] The surface point cloud of metal parts is divided into regions by using the density-based spatial clustering of applications with noise (DBSCAN) algorithm to complete the automatic detection of crack defects and obtain the point cloud corresponding to the crack defects.

[0014] Furthermore, in step (3), a Cartesian coordinate system is established with the surface of the metal part where the defective area is located as the XY plane and the positive direction of the Z axis perpendicular to the surface upward; the point cloud corresponding to the dent or peeling block defect is projected onto the XY plane, and the distance between each point in the point cloud before and after the projection is calculated;

[0015] The point cloud corresponding to the dent or peeling defect is processed in layers; the thickness h of each layer of point cloud is compared with the minimum z coordinate z of the midpoint of the point cloud corresponding to the defect. min Define the minimum z coordinate z of the kth layer of point cloud from bottom to top k,min for:

[0016] z k,min =z min+h(k-1)

[0017] Traverse each point in the point cloud corresponding to the dent or peeling block defect. If the z coordinate of a point belongs to the interval [z k,min , z k,min +h), then the point is determined to belong to the kth layer of point cloud; after completing the layered processing of the point cloud corresponding to the dent or peeling defect, the upper boundary contour of each layer of point cloud is obtained based on the convex hull (CH) algorithm, and a series of bias curves are filled in the upper boundary contour to generate the contour-parallel laser scanning path of each layer of point cloud, or a series of parallel line segments are filled in the upper boundary contour to generate the direction-parallel laser scanning path of each layer of point cloud.

[0018] Furthermore, in step (3), the cleaning laser head performs laser cleaning on the wall surface of the concave or peeling defect area to remove rust, oil stains, etc. on the wall surface, thereby reducing the impact on the quality of laser additive repair; the movement path of the cleaning laser head in the XY plane is determined by the laser scanning path of the top layer of the point cloud, the movement path of the cleaning laser head in the Z direction is determined by the distance between the front and back projections of each point in the point cloud, and the cleaning laser incident angle is determined by the normal vector of each point in the point cloud; the additive repair laser head performs laser additive repair on the concave or peeling defect, and the movement path of the additive repair laser head is determined by the laser scanning path of each layer of the point cloud.

[0019] Furthermore, in step (4), for the point cloud corresponding to the crack defect, the outer contour of the crack defect is extracted based on the CH algorithm, and parallel line segments are filled in the contour to generate a laser scanning path; the cleaning laser head performs laser cleaning on the crack defect to remove rust, oil stains, etc. in the crack defect area to improve the repair quality; the moving path of the cleaning laser head is determined by the laser scanning path; the additive repair laser head performs laser additive repair on the crack defect, and the moving path of the additive repair laser head is determined by the laser scanning path.

[0020] Compared with the prior art, the above technical solution conceived by the present invention has the following beneficial effects:

[0021] 1. The present invention proposes an automatic detection and laser additive repair method for surface defects of metal parts. The method obtains a point cloud of the metal part surface by scanning with a scanner, and automatically detects dents, peeling and crack defects on the metal part surface by analyzing the point cloud features. There is no need to locate and classify the defects in advance, and the method has high efficiency.

[0022] 2. The present invention proposes a method for automatic detection and laser additive repair of surface defects of metal parts. Before laser additive repair, the defects are laser cleaned to remove rust, oil stains, etc. in the defective area, thereby improving the quality of laser additive repair.

[0023] 3. The present invention proposes an automatic detection and laser additive repair method for surface defects of metal parts, which processes the point cloud corresponding to the defect, generates a laser scanning path in the defect area, and performs laser additive repair on the defect. There is no need to perform manual measurement, modeling, etc. on the defects in advance, thereby improving the repair efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A schematic diagram of automatic detection and laser additive repair of surface defects of metal parts according to the present invention;

[0025] Figure 2 A schematic diagram of a point cloud of a metal part surface obtained by scanning with a scanner of the present invention;

[0026] Figure 3 This is a schematic diagram of a metal part surface point cloud obtained after preprocessing of the present invention;

[0027] Figure 4 A schematic diagram of a point cloud corresponding to a concave defect of the present invention;

[0028] Figure 5 It is a schematic diagram of point clouds of each layer corresponding to the concave defect of the present invention;

[0029] Figure 6 It is a schematic diagram of the upper boundary contour of each layer of point cloud corresponding to the concave defect of the present invention;

[0030] Figure 7 It is a schematic diagram of the laser scanning path of each layer of point cloud of the present invention;

[0031] Figure 8 It is a schematic diagram of the laser cleaning process of the present invention;

[0032] Fig. 9 It is a schematic diagram of the laser additive repair process of the present invention;

[0033] In the figure: 1-metal part, 2-scanner, 3-cleaning laser head, 4-additive repair laser head, 5-defect, 6-upper boundary contour of each layer of point cloud corresponding to the concave defect, 7-laser scanning path of each layer of point cloud, 8-repairing layer, 501-concave defect, 502-peeling defect, 503-crack defect, 701-laser scanning path of the top layer of point cloud. DETAILED DESCRIPTION

[0034] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, the following will be described in detail with reference to the accompanying drawings and specific embodiments. The exemplary embodiments of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0035] A method for automatically detecting and repairing surface defects of metal parts by laser additive manufacturing, wherein a scanner 2 is used to scan the surface of a metal part 1, and the surface point cloud features of the metal part 1 obtained by scanning are analyzed to automatically detect a defect 5 on the surface of the metal part 1, wherein the defect 5 includes a concave defect 501, a peeling defect 502, and a crack defect 503. The point cloud corresponding to the defect 5 is processed, and a moving path of a cleaning laser head 3 and an additive manufacturing repair laser head 4 is generated in the defect 5 area. The cleaning laser head 3 performs laser cleaning on the defect 5, and the additive manufacturing repair laser head 4 performs laser additive manufacturing repair on the defect 5 to form a repair layer 8, such as Figure 1 shown.

[0036] In the following embodiments, the metal part 1 is a U75V rail, and the upper surface concave defect 501 is taken as an example to illustrate the specific implementation of the method. The depth of the concave defect 501 is 9.8 mm.

[0037] Scan the surface of the metal part 1 where the concave defect 501 is located by using a scanner 2 to obtain a point cloud of the surface of the metal part 1, such as Figure 2 shown.

[0038] The surface point cloud of metal part 1 is preprocessed; the point cloud is denoised based on the statistical outlier removal (SOR) algorithm, the average distance between each point in the point cloud and its 20 neighboring points is calculated, and each point in the point cloud is traversed. Points whose average distance is not within the normal distribution interval are determined as noise points and removed to complete the denoising of the point cloud; the point cloud is downsampled based on the voxel grid filtering (VGF) algorithm, and the cube with the smallest volume that can contain the point cloud is calculated. The cube is divided into multiple cubic voxels. For the cubic voxel containing the point, the centroid of the voxel is used to replace the point contained in the voxel to form a new point cloud, and the downsampling of the point cloud is completed; the surface point cloud of metal part 1 obtained after preprocessing is as follows Figure 3 shown.

[0039] The concave defect 501 on the surface of the metal part 1 is automatically detected by analyzing the normal vector of each point in the point cloud of the metal part 1 obtained after preprocessing; the normal vector of each point in the point cloud is calculated based on the KNN algorithm, and the local plane in the least squares sense of each point in the point cloud and its 30 neighboring points is calculated, and the normal vector of the local plane is the normal vector of each point in the point cloud; the specific algorithm for automatically detecting the concave defect 501 on the surface of the metal part 1 by analyzing the normal vector of each point in the point cloud is:

[0040] Define the normal vector P of any point P in the point cloud P The average of the absolute values ​​of the differences between the normal vectors of its 50 neighboring points for:

[0041]

[0042] like If the value is greater than the set threshold value 0.01, then point P is determined to belong to the concave defect 501; traverse each point in the point cloud, select all points belonging to the concave defect 501, complete the automatic detection of the concave defect 501, and obtain the point cloud corresponding to the concave defect 501, such as Figure 4 shown.

[0043] During the laser cleaning process, the cleaning laser head 3 needs to move in the Z direction to ensure that the vertical distance between the cleaning laser head 3 and the wall of the recessed defect 501 area remains unchanged; the point cloud corresponding to the recessed defect 501 is projected to the surface of the metal part 1 where the recessed defect 501 area is located, and the distance between each point in the point cloud before and after the projection is calculated, and the moving path of the cleaning laser head 3 in the Z direction is determined using this distance.

[0044] The surface of the metal part 1 where the concave defect 501 is located is taken as the XY plane, and the positive direction of the Z axis is perpendicular to the surface and upward. A Cartesian coordinate system is established, and the point cloud corresponding to the concave defect 501 is projected onto the XY plane. The specific algorithm of the projection is:

[0045] Assume that the coordinates of the i-th point in the point cloud corresponding to the concave defect 501 are (x i ,y i ,z i ), the parameter t required for projecting the i-th point in the point cloud onto the XY plane is:

[0046]

[0047] The coordinates of the i-th point in the point cloud after being projected onto the XY plane (x i ′,y i ′,z i ')for:

[0048] x i ′=0×t+x i

[0049] y i ′=0×t+y i

[0050] z i ′=1×t+z i

[0051] Calculate the distance between the projection of each point in the point cloud corresponding to the concave defect 501 and the distance d between the projection of the i-th point in the point cloud i for:

[0052]

[0053] For the point cloud corresponding to the concave defect 501 , a laser scanning path is generated in the concave defect 501 area through steps such as point cloud layering processing, upper boundary contour acquisition, and offset curve filling.

[0054] The point cloud corresponding to the concave defect 501 is layered, and the layering algorithm is specifically as follows:

[0055] The thickness of each layer of point cloud is set to 0.7 mm, and the minimum z coordinate of the midpoint of the point cloud corresponding to the concave defect 501 is z min is -9.8 mm, the minimum z coordinate z of the kth layer of point cloud from bottom to top k,min for:

[0056] z k,min =-9.8+0.7×(k-1)

[0057] Traverse each point in the point cloud corresponding to the concave defect 501. If the z coordinate of a point belongs to the interval [z k,min , z k,min +h), then it is determined that the point belongs to the kth layer of point cloud; the layered processing of the point cloud corresponding to the concave defect 501 is completed to obtain the point clouds of each layer corresponding to the concave defect 501, such as Figure 5 shown.

[0058] Based on the CH algorithm, the upper boundary contour 6 of each layer of point cloud corresponding to the concave defect 501 is obtained, and the minimum two-dimensional convex polygon parallel to the XY plane that can completely contain each layer of point cloud is found, and the two-dimensional convex polygon is regarded as the upper boundary contour 6 of each layer of point cloud, such as Figure 6 shown.

[0059] A series of bias curves are filled in the upper boundary contour 6 of each layer of point cloud corresponding to the concave defect 501 to generate the laser scanning path 7 of each layer of point cloud, such as Figure 7 shown.

[0060] The moving path of the cleaning laser head 3 in the XY plane is determined by the laser scanning path 701 of the topmost point cloud in the laser scanning paths 7 of each layer of the point cloud. The moving path of the cleaning laser head 3 in the Z direction is determined by the distance between the projections of each point in the point cloud. The cleaning laser incident angle is determined by the normal vector of each point in the point cloud. The laser cleaning process parameters are set as laser power 500W and scanning speed 6000mm / s. The cleaning laser head 3 performs laser cleaning on the wall surface of the concave defect 501 area on the surface of the metal part 1 according to the determined moving path, cleaning laser incident angle and laser cleaning process parameters, such as Figure 8 As shown; after the laser head 3 to be cleaned completes scanning along the determined moving path, the rust, oil stains, etc. on the wall surface of the concave defect 501 area are removed.

[0061] The moving path of the additive repair laser head 4 is determined by the laser scanning path 7 of each layer of the point cloud; the laser additive repair process parameters are set as laser power 2000W, scanning speed 4mm / s, and powder feeding speed 15g / min; the powder material is Fe-58, and the particle size range is 80-180 mesh; the additive repair laser head 4 performs laser additive repair on the concave defect 501 on the surface of the metal part 1 according to the determined moving path, laser additive repair process parameters and powder material to form a repair layer 8, as shown in FIG. Fig. 9 As shown; after the additive repair laser head 4 completes scanning along the determined moving path, the repair of the concave defect 501 is completed.

[0062] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for automatic detection and laser additive repair of surface defects of metal parts, which uses a scanner to obtain a point cloud of the metal part surface, automatically detects depressions, peeling and crack defects on the metal part surface by analyzing the point cloud, generates a laser scanning path in the defect area, and performs laser cleaning and laser additive repair on the defect, characterized in that: The method comprises the following steps: (1) Scan the surface of metal parts with a scanner and pre-process the obtained point cloud; (2) Automatically detect dents and peeling defects on the surface of metal parts by analyzing the normal vectors of each point in the point cloud, and automatically detect crack defects on the surface of metal parts by analyzing the point cloud density; (3) Performing layered processing on the point cloud corresponding to the dent or peeling defect, generating a laser scanning path in the dent or peeling defect area, and performing laser cleaning and laser additive repair on the defect; (4) Based on the point cloud corresponding to the crack defect, a laser scanning path is generated in the crack defect area, and the crack defect is laser cleaned and laser additively repaired.

2. The method for automatic detection and laser additive repair of metal part surface defects according to claim 1, characterized in that: In step (2), the normal vector of each point in the surface point cloud of the metal part is calculated based on the K-Nearest Neighbor (KNN) algorithm; the normal vector P of any point P in the point cloud is defined as P The average of the absolute values ​​of the differences between the normal vectors of other points in its neighborhood for: like If it is greater than the set threshold, then point P is determined to be a dent or peeling defect; traverse all points in the point cloud, filter out all points that belong to dents or peeling defects, complete automatic detection of dents or peeling defects, and obtain the point cloud corresponding to the dent or peeling defect; The surface point cloud of metal parts is divided into regions by using the density-based spatial clustering of applications with noise (DBSCAN) algorithm to complete the automatic detection of crack defects and obtain the point cloud corresponding to the crack defects.

3. The method for automatic detection and laser additive repair of metal part surface defects according to claim 1, characterized in that: In step (3), a Cartesian coordinate system is established with the surface of the metal part where the defective area is located as the XY plane and the positive direction of the Z axis perpendicular to the surface upward; the point cloud corresponding to the dent or peeling defect is projected onto the XY plane, and the distance between each point in the point cloud before and after the projection is calculated; The point cloud corresponding to the dent or peeling defect is processed in layers; the thickness h of each layer of point cloud is compared with the minimum z coordinate z of the midpoint of the point cloud corresponding to the defect. min Define the minimum z coordinate z of the kth layer of point cloud from bottom to top k,min for: z k,min =z min +h(k-1) Traverse each point in the point cloud corresponding to the dent or peeling block defect. If the z coordinate of a point belongs to the interval [z k,min , z k,min +h), then the point is determined to belong to the kth layer of point cloud; after completing the layered processing of the point cloud corresponding to the dent or peeling block defect, the upper boundary contour of each layer of point cloud is obtained based on the convex hull (CH) algorithm, and a series of bias curves are filled in the upper boundary contour to generate the contour-parallel laser scanning path of each layer of point cloud, or a series of parallel line segments are filled in the upper boundary contour to generate the direction-parallel laser scanning path of each layer of point cloud.

4. The method for automatic detection and laser additive repair of metal part surface defects according to claim 1, characterized in that: In step (3), the cleaning laser head performs laser cleaning on the wall surface of the concave or peeling defect area to remove rust, oil stains, etc. on the wall surface, thereby reducing the impact on the quality of laser additive repair; the movement path of the cleaning laser head in the XY plane is determined by the laser scanning path of the top layer of the point cloud, the movement path of the cleaning laser head in the Z direction is determined by the distance between the projections of each point in the point cloud, and the incident angle of the cleaning laser is determined by the normal vector of each point in the point cloud; the additive repair laser head performs laser additive repair on the concave or peeling defect, and the movement path of the additive repair laser head is determined by the laser scanning paths of each layer of the point cloud.

5. The method for automatic detection and laser additive repair of metal part surface defects according to claim 1, characterized in that: In step (4), for the point cloud corresponding to the crack defect, the outer contour of the crack defect is extracted based on the CH algorithm, and parallel line segments are filled in the contour to generate a laser scanning path; the cleaning laser head performs laser cleaning on the crack defect to remove rust, oil stains, etc. in the crack defect area to improve the repair quality; the moving path of the cleaning laser head is determined by the laser scanning path; the additive repair laser head performs laser additive repair on the crack defect, and the moving path of the additive repair laser head is determined by the laser scanning path.

Images