Water turbine runner crack electric arc additive repair trajectory planning method based on irregular model

Abstract

The invention provides a water turbine runner crack electric arc additive repairing track planning method based on an irregular model, and relates to the technical field of hydroelectric power generation equipment crack repairing. The method comprises the steps that crack defects are removed through carbon arc air gouging, and a model of a blade area to be repaired is established; based on a reverse reconstruction technology and a Boolean operation technology, generating a three-dimensional grid model of the irregular repair area; a mixed trajectory planning strategy is optimized, and the repairing area is sliced to generate an electric arc additive repairing trajectory; and welding gun posture optimization based on collision detection is carried out, and the contour track is adjusted according to a given rule. A geometric model of a to-be-repaired area of a blade is established, and a three-dimensional grid model of an irregular repaired area is quickly and accurately established based on a reverse reconstruction technology and Boolean operation.

Description

Technical Field

[0001] The present invention relates to the technical field of crack repair of hydroelectric power generation equipment, and particularly relates to a method for planning the arc additive repair trajectory of a water turbine runner crack based on an irregular model. Background Art

[0002] The water turbine runner is the core component of a hydropower station, and its operation directly affects the power generation efficiency and the long-term reliability of the equipment. The runner is usually made of high-strength alloy materials (such as stainless steel, high manganese steel, etc.), and has a complex geometric shape and a large volume. Due to long-term operation under high water flow impact, alternating loads, and corrosive environments, the runner is prone to damage such as cracks. If the cracks are not repaired in time, it may reduce the hydraulic efficiency of the water turbine or even lead to equipment shutdown or serious accidents. Traditional repair methods include carbon arc air gouging to remove cracks, welding repair, and subsequent grinding and polishing. Manual welding repair may produce secondary defects such as residual stress, pores, and cracks, and it is difficult to accurately repair in complex and irregular crack areas, and the repair quality depends on manual experience. In welding repair technology, wire and arc additive manufacturing (WAAM) directly repairs the crack area by layer-by-layer deposition of metal materials. By using digital modeling and trajectory planning technologies, it can accurately repair irregular crack areas with complex shapes. In order to improve the efficiency and quality of runner crack repair and deal with complex-shaped irregular crack areas, it is particularly important to study the arc additive repair trajectory planning technology for water turbine runner cracks.

[0003] However, there are the following difficulties in the arc additive repair of water turbine runner cracks. First, the geometric features of the area to be repaired of the runner crack are relatively complex, and it is difficult to accurately extract the repair area. Second, the crack distribution of the runner is random and complex, and the traditional hybrid trajectory planning strategy is difficult to adapt to the irregular shape of the crack, and overfilling or non-filling phenomena are likely to occur. Finally, for the WAAM technology, the welding torch usually keeps perpendicular to the substrate surface to ensure good deposition quality. However, the runner surface has the characteristics of a three-dimensional curved surface, and the crack area may have large curvature changes and irregular inclinations, and it is easy to collide with the mold in some areas. If the posture of the welding torch is not adjusted, it will hinder the repair process. In view of the above difficulties, a method for planning the additive repair trajectory of an irregular crack area is studied. Through reverse reconstruction technology and Boolean operation technology, a three-dimensional grid model of the repair area is accurately generated; combined with a hybrid trajectory planning strategy, the trajectory generation is optimized to adapt to complex crack areas, and overfilling or non-filling phenomena during the additive process are reduced or even eliminated; based on collision detection, the contour trajectory is adjusted according to a given rule, and the posture of the welding torch is optimized, ensuring the stability of the welding process and the repair quality. Summary of the Invention

[0004] The object of the present invention is to provide a method for planning the arc additive repair trajectory of a crack on a water turbine runner based on an irregular model. This method quickly and accurately establishes a three-dimensional grid model of the irregular repair area by establishing a geometric model of the blade area to be repaired and based on reverse reconstruction technology and Boolean operations; through a hybrid trajectory planning strategy and a welding torch attitude optimization algorithm, the repair area is sliced to generate an arc additive repair trajectory and the attitude of the welding torch is adjusted. Through the arc additive repair trajectory planning method based on an irregular model, accurate identification of the crack area and generation of a highly adaptable trajectory can be achieved, significantly improving the repair quality and efficiency.

[0005] In order to achieve the above technical features, the object of the present invention is realized as follows: A method for planning the arc additive repair trajectory of a crack on a water turbine runner based on an irregular model, comprising the following steps: S1. Remove the crack defect by carbon arc air gouging and establish a model of the blade area to be repaired; S2. Generate a three-dimensional grid model of the irregular repair area based on reverse reconstruction technology and Boolean operation technology; S3. Optimize the hybrid trajectory planning strategy and slice the repair area to generate an arc additive repair trajectory; S4. Optimize the welding torch attitude based on collision detection and adjust the contour trajectory according to the given rules.

[0006] Preferably, the specific steps in S1 include the following: S11. Detect and evaluate the crack on the blade surface to clarify the scope and depth of the defect; S12. Use carbon arc air gouging to remove the crack and defective materials, move the electrode along the crack direction, and melt and blow away the materials in the crack and its surrounding fatigue area; S13. Use a grinding wheel or polishing tool to trim the surface after air gouging, remove the oxide layer and slag, and ensure that the surface is smooth and free of secondary defects; S14. Use a 3D scanner to accurately scan the blade surface after air gouging to obtain high-resolution point cloud data.

[0007] Preferably, the detection methods for detecting the crack on the blade surface in S11 include: penetrant testing and ultrasonic testing means.

[0008] Preferably, during the accurate scanning of the blade surface after air gouging in S14, ensure that the scanning range includes the complete geometric information of the removal area and its surroundings.

[0009] Preferably, the specific steps in S2 include the following: S21. Perform surface fitting and modeling on the point cloud processed in step S1. Use the least squares method to fit the surface of the crack area and generate a smooth NURBS surface. Convert the fitted surface into a triangulated mesh to form a preliminary model of the crack gouging area. S22. Adopt Boolean operation technology to combine the crack area model with the original design model, and generate an accurate three-dimensional model of the repair area through geometric operations. After the three-dimensional mesh model of the repair area is generated through Boolean operations, it needs to be further optimized to ensure its applicability to the additive repair trajectory planning. Among them, the optimization process includes: removing non-manifold edges, duplicate patches or isolated points generated during Boolean operations; densifying the mesh to accurately describe complex geometric features and ensure that the mesh resolution of the crack repair area meets the requirements of the additive repair trajectory.

[0010] Preferably, the triangulated mesh in S21 adopts the STL format. The specific acquisition process of the accurate three-dimensional model of the repair area in S22 includes: In the CAD modeling software, align the crack area model after gouging with the complete blade model. First, offset the surface of the designed blade model in the direction of its normal by a certain distance. Subsequently, outline the area to be repaired on the crack area model. Finally, construct the offset surface within the circled area into a solid, and perform a Boolean subtraction operation between this solid and the crack area model to obtain an accurate model of the area to be repaired.

[0011] Preferably, S3 specifically includes the following steps: S31. Process the unfilled area caused by contour line offset: Calculate the possible unfilled area during the offset process through the contour line offset algorithm. First, offset the model contour line lines1 inward at a given interval to obtain the contour trajectory line lines2. Secondly, offset the obtained contour trajectory line outward by the same distance, denoted as the model contour line lines3. Finally, perform a Boolean operation on lines1 and lines3. When the difference set between the two exists and the area is greater than a certain threshold, it is regarded as a valid area for filling. S32. Determine the direction of the scan line: Design a method for selecting the scan line direction to determine the direction of the scan line.

[0012] Preferably, the method for selecting the scan line direction in S32 is specifically as follows: The inclination angle θ of the scan line with the positive direction of the X-axis represents the direction of the scan line. First, calculate the length of each line segment in the contour line and its angle with the X-axis, and the angle range is 0° to 180°, which is evenly divided into 6 intervals. Secondly, calculate the total length of the line segments in each interval. Finally, select the interval with the smallest total length of the line segments, and the inclination angle θ is equal to the median value of this interval. Among them, when there is only one line segment in the selected interval, the inclination angle θ is equal to the angle between the line segment and the X-axis.

[0013] Preferably, step S4 specifically includes the following steps: S41. Propose two preconditions to optimize the collision detection algorithm: First, simplify the welding torch into a cylinder for collision detection with the mold. Assume that during the entire repair process, the nozzle at the tail of the welding torch colliding with the mold is the main calculation object. Second, when generating the trajectory, ensure that the distance between adjacent target points in the trajectory is within a certain range. Based on the above assumptions, when the welding torch does not collide with the mold at the positions of two target points, the welding torch will not collide with the mold at any position on the trajectory between the two target points. Furthermore, at each discrete point, convert the dynamic collision detection process into a static collision detection process; S42. Preliminary detection and precise detection: In the preliminary detection stage, divide the mold into a certain number of axis-aligned bounding boxes through space partitioning, and then perform collision detection between the bounding boxes and the cylinder. In the precise detection stage, perform collision detection between the triangular patches and the cylinder; S43. The arc additive manufacturing for repairing the runner crack adopts a contour trajectory and a zigzag trajectory.

[0014] Preferably, step S43 specifically includes: Optimize the posture of the contour trajectory. When the welding torch in the upright posture collides, rotate the posture of the welding torch clockwise by a certain angle around the forward direction, and the adjustment angle cannot exceed the preset maximum adjustment angle. If the welding torch cannot avoid colliding with the repair area, then adjust the forward direction and perform a new round of detection; the adjustment amount of the forward direction is optional and limited.

[0015] The present invention has the following beneficial effects: 1. Deal with the unfilled area caused by the offset of the contour line. The sharp corners on the trajectory will cause serious forming problems. When offsetting the contour line at the sharp corners, an unfilled phenomenon will occur at the sharp corner positions. Through the method provided by the present invention, it is possible to identify the unfilled areas existing in the process of contour line offset and determine the unfilled areas according to the area threshold.

[0016] 2. Determine the optimal scan line direction for filling the trajectory. The hybrid trajectory generation algorithm first determines the relatively optimal scan line direction according to the inclination angle interval with the minimum total length of the contour line, which can greatly reduce the appearance of sharp angles.

[0017] 3. Avoid the collision between the welding torch and the area to be repaired on the runner by adjusting the posture of the welding torch. Determine whether the welding torch collides with the runner model through the collision detection algorithm based on space partitioning. If there is no collision, the upright posture is the posture of the welding torch at the current point. If a collision occurs, adjust the posture of the welding torch according to the predetermined rules and repeat the detection until there is no collision. Adjusting the posture of the welding torch ensures the stability of the welding process and the repair quality. Description of the Drawings

[0018] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0019] Figure 1 It is a specific flowchart for the implementation of the present invention.

[0020] Figure 2 It is a runner blade diagram of the present invention.

[0021] Figure 3 It is an effect diagram of carbon arc air gouging of the present invention.

[0022] Figure 4 It is an effect diagram of rough grinding with a grinding wheel of the present invention.

[0023] Figure 5 It is a scanned point cloud data diagram of the present invention.

[0024] Figure 6 It is a model reconstruction result diagram of the present invention.

[0025] Figure 7 It is a repaired area extraction diagram of the present invention.

[0026] Figure 8 It is an optimized repair process diagram of the present invention.

[0027] Figure 9 It is a refined grid process diagram of the present invention.

[0028] Figure 10 It is a schematic diagram of the unfilled area with sharp corners of the present invention.

[0029] Figure 11 It is an effect diagram of filling sharp corners with contour offset of the present invention.

[0030] Figure 12 It is a scanning direction diagram of the present invention.

[0031] Figure 13 It is a statistical chart of the total length of line segments within each corner interval of the present invention.

[0032] Figure 14 (a) and (b) are the 45° and 135° scanning directions of the present invention.

[0033] Figure 15 It is a welding torch and repaired model diagram of the present invention.

[0034] Figure 16 It is a spatial division diagram of the blade model of the present invention.

[0035] Figure 17 It is a contour trajectory adjustment strategy diagram of the present invention.

[0036] Figure 18 It is a contour trajectory adjustment result diagram of the present invention. Specific implementation manners

[0037] The following further describes the implementation manners of the present invention in conjunction with the accompanying drawings.

[0038] Embodiment 1: Referring to Figure 1 , a method for planning the arc additive repair trajectory of the crack of a hydraulic turbine runner based on an irregular model, includes the following steps: S1. Remove the crack defect by carbon arc gouging and establish a model of the blade area to be repaired: S11. Detect and evaluate the crack on the blade surface to clarify the scope and depth of the defect. Common detection methods include: means such as penetrant testing and ultrasonic testing.

[0039] S12. Use carbon arc gouging to efficiently remove the crack and defective materials. Move the operating electrode along the crack direction to melt and blow away the materials in the crack and its surrounding fatigue area.

[0040] S13. Use a grinding wheel or polishing tool to trim the surface after gouging to remove the oxide layer and slag, ensuring that the surface is smooth and free of secondary defects.

[0041] S14. Use a 3D scanner to accurately scan the blade surface after gouging to obtain high-resolution point cloud data. Ensure that the scanning range includes the complete geometric information of the removed area and its surroundings.

[0042] S2. Generate a three-dimensional mesh model of the irregular repair area based on reverse reconstruction technology and Boolean operation technology: S21. Perform surface fitting and modeling on the point cloud processed in step S1. Use the least squares method to fit the crack area surface to generate a smooth NURBS surface. Convert the fitted surface into a triangulated mesh (such as STL format) to form a preliminary crack gouging area model.

[0043] S22. Combine the crack area model with the original design model by Boolean operation technology to generate an accurate three-dimensional model of the repair area through geometric operations. In the CAD modeling software, align the crack gouging area model after gouging with the complete blade model. First, offset the surface of the designed blade model by a certain distance in the direction of its normal. Subsequently, outline the area to be repaired on the crack area model. Finally, construct the offset surface within the circled area into a solid, and perform a Boolean subtraction operation on this solid and the crack area model to obtain an accurate model of the area to be repaired.

[0044] After the 3D mesh model of the repair area is generated through Boolean operations, it needs to be further optimized to ensure its applicability to additive repair trajectory planning. This mainly includes: removing non-manifold edges, duplicate faces, or isolated points generated during Boolean operations. Refining the mesh to accurately describe complex geometric features and ensuring that the mesh resolution of the crack repair area meets the requirements of the additive repair trajectory.

[0045] S3. Optimize the hybrid trajectory planning strategy and slice the repair area to generate the arc additive repair trajectory: S31. Process the unfilled area caused by contour offset. Sharp corners on the trajectory can cause serious forming problems. When offsetting the contour line at a sharp corner, an unfilled phenomenon will occur at the sharp corner position. Calculate the possible unfilled area during the offset process through the contour offset algorithm. First, offset the model contour line lines1 inward by a given interval to obtain the contour trajectory line lines2. Secondly, offset the obtained contour trajectory line outward by the same distance, denoted as the model contour line lines3. Finally, perform a Boolean operation on lines1 and lines3. When the difference set between the two exists and its area is greater than a certain threshold, it is regarded as a valid area for filling.

[0046] S32. Determine the direction of the scan line. When the angle between the scan line and the contour line is small, sharp corners usually appear. Therefore, finding the relatively optimal scan line direction helps to greatly reduce the possibility of sharp corners. Design a method for selecting the scan line direction. The inclination angle θ of the scan line with the positive X-axis direction represents the direction of the scan line. First, calculate the length of each line segment in the contour line and its angle with the X-axis. The angle range is 0° to 180°, which is evenly divided into 6 intervals. Secondly, calculate the total length of the line segments in each interval. Finally, select the interval with the smallest total length of line segments, and the inclination angle θ is equal to the median value of this interval. In particular, when there is only one line segment in the selected interval, the inclination angle θ is equal to the angle between the line segment and the X-axis.

[0047] S4. Optimize the torch posture based on collision detection and adjust the contour trajectory according to the given rules: S41. Propose two preconditions to optimize the collision detection algorithm. First, the torch can be simplified to a cylinder for collision detection with the mold. Assume that during the entire repair process, the nozzle at the tail of the torch colliding with the mold is the main calculation object. Second, when generating the trajectory, ensure that the distance between adjacent target points in the trajectory is not too far. This allows the assumption that when the torch does not collide with the mold at the positions of two target points, the torch will not collide with the mold at any position on this trajectory between the two target points. Therefore, convert the dynamic collision detection process into a static collision detection process at each discrete point.

[0048] S42. Preliminary detection and precise detection. In the preliminary detection stage, the mold is divided into a certain number of axis pair bounding boxes through space division, and then collision detection is performed between the bounding boxes and the cylinder. In the precise detection stage, collision detection is performed between the triangular facets and the cylinder.

[0049] S43. The arc additive manufacturing for repairing the runner crack adopts a contour trajectory and a zigzag trajectory. The attitude of the contour trajectory is optimized. When the welding torch in the upright posture collides, the attitude of the welding torch is rotated clockwise by a certain angle around the advancing direction, and the adjustment angle cannot exceed the preset maximum adjustment angle. If the welding torch cannot avoid colliding with the repair area, the advancing direction is adjusted, and a new round of detection is performed. The adjustment amount of the advancing direction is optional and limited.

[0050] Embodiment 2: As Figure 1 shown, a method for planning the arc additive repair trajectory of a hydraulic turbine runner crack based on an irregular model includes the following steps: Please refer to Figures 1 - 18 , the embodiments of the present invention provide a method for planning the arc additive repair trajectory of a hydraulic turbine runner crack based on an irregular model, which is applied to the repair work of hydraulic turbine runner cracks and mainly includes the following steps: S1. Use carbon arc air gouging to remove crack defects and establish a model of the blade area to be repaired. Specifically: S11. Refer to Figure 2 , taking the Francis turbine runner of a certain hydropower station as the research object, detect and evaluate the cracks on the blade surface to clarify the scope and depth of the defects. The detection method used is penetrant testing (PT), and surface cracks are detected through capillary action. Evaluate the cracks on the runner blade through PT testing. The length of the blade crack is about 130 mm, and the distance from the water outlet edge of the blade is about 670 mm.

[0051] S12. Use carbon arc air gouging to efficiently remove the crack and defective materials. Move the operating electrode along the crack direction to melt and blow away the materials in the crack and its surrounding fatigue area. Carbon arc air gouging uses the high temperature generated by the arc on the workpiece surface to melt the metal, and at the same time uses a high-speed air flow to blow the melted metal away from the working surface. Select the current intensity, gas flow rate and gouging speed according to the material type and defect depth. The current intensity is set to 350 A, and the gas is high-pressure compressed air. Ensure that the gouging depth is greater than the crack depth to completely remove the crack. For the specific effect after carbon arc air gouging, please refer to Figure 3 .

[0052] S13. Use a grinding wheel or polishing tool to trim the surface after air gouging to remove the oxide layer and slag, and ensure that the surface is smooth and free of secondary defects. For the specific effect after rough grinding with a grinding wheel, please refer to Figure 4 .

[0053] S14. Use a 3D scanner to precisely scan the surface of the blade after air gouging to obtain high-resolution point cloud data. Ensure that the scanning range includes the complete geometric information of the removed area and its surroundings. Use software (such as Geomagic, PolyWorks) to denoise and smooth the point cloud data. For specific details of the scanned point cloud data, refer to Figure 5 。

[0054] S2. Based on reverse reconstruction technology and Boolean operation technology, generate a three-dimensional mesh model of the irregular repair area. Specifically: S21. The point cloud processed in S1 can be used for surface fitting and modeling. Use the Squared Distance Minimization (SDM) method based on the least squares method to fit the surface of the crack area and generate a smooth NURBS surface. Utilize NURBS surface discretization technology to convert the fitted surface into a triangulated mesh (such as STL format) to form a preliminary crack area model. For specific results of model reconstruction, refer to Figure 6 。

[0055] S22. The Boolean operation technology combines the crack area model with the original design model to generate an accurate three-dimensional model of the repair area through geometric operations. In CAD modeling software, align the model to be repaired after air gouging with the complete blade model. After the two models are aligned, the Boolean operation cannot be directly executed. This is because problems such as scanning errors, positioning errors, and thermal expansion and contraction are inevitable, resulting in unsatisfactory dimensional accuracy of the blade model to be repaired. In addition, directly performing the Boolean operation will cause many isolated patch features to appear on the model to be repaired. Therefore, first offset the surface of the designed blade model in the direction of its normal by a certain distance. Subsequently, outline the area to be repaired on the crack area model. Finally, construct the offset surface within the circled area into a solid and perform a Boolean subtraction operation between this solid and the crack area model to obtain an accurate model of the area to be repaired. For specific results of repair area extraction, refer to Figure 7 。

[0056] After the three-dimensional mesh model of the repair area is generated through Boolean operations, it needs to be further optimized to ensure its suitability for additive repair trajectory planning. Refer to Figure 8 ,The optimization of the repair process is as follows: First, optimize the mesh features to remove non-manifold edges, duplicate patches, or isolated points generated during the Boolean operation.

[0057] Refer to Figure 9 ,Then perform mesh subdivision to accurately describe complex geometric features and ensure that the mesh resolution of the crack repair area meets the requirements of the additive repair trajectory. Use the classic Loop mesh subdivision algorithm to refine the mesh resolution to capture the complex geometric features of the crack repair area.

[0058] S3. Optimize the hybrid trajectory planning strategy to slice the repair area and generate the arc additive repair trajectory. Specifically: S31. Process the unfilled area caused by contour line offset. Refer to Figure 10 , the sharp corners on the trajectory will cause serious forming problems. When offsetting the contour line at the sharp corners, unfilled phenomena will occur at the sharp corner positions.

[0059] To calculate the unfilled area that may occur during the contour line offset process, the following steps are used for analysis and processing: First, according to the initial contour line lines1 of the model, offset it inward at a set interval to generate the offset contour trajectory line lines2. Then, offset the generated contour trajectory line lines2 outward by the same distance to restore it to another set of contour lines lines3. Finally, compare the original contour line lines1 and the restored contour line lines3 through Boolean operations. If the difference set between the two exists and the area of the difference set exceeds the set threshold, the difference set is determined as the unfilled area and filled. Refer to Figure 11 , for the filling effect of the contour offset sharp corners.

[0060] S32. Determine the direction of the scan line. According to the requirements of the WAAR hybrid trajectory planning, after implementing the contour trajectory, internal trajectory filling is required. After slicing the model with a predetermined layer thickness, the contour lines of each layer are obtained. Obviously, the contour trajectory consists of many straight line segments. The internal zigzag trajectory is generated by the intersection of the scan line and the offset contour line. At this stage, the direction of the scan line has a great influence on the generated trajectory. It will affect the size of the unfilled and overfilled areas at the trajectory corners. However, due to the complex and irregular contour shape, it is difficult to find an optimal direction to minimize the overfilled or unfilled area.

[0061] Only the sharp corners on the trajectory will cause serious forming problems. Sharp corners usually occur when the angle between the scan line and the contour line is small. Therefore, finding a relatively optimal scan line direction can greatly reduce the possibility of sharp corners.

[0062] Design a method to select the scan line direction. The inclination angle θ of the scan line with the positive x-axis direction represents the direction of the scan line. The method is as follows: First, calculate the length of each line segment in the contour line and its angle with the x-axis. The angle range is 0° to 180°, which is evenly divided into 6 intervals. Second, calculate the total length of the line segments in each interval. Finally, select the interval with the smallest total length of the line segments, and the inclination angle θ is equal to the median of this interval. Refer to Figure 12 , in particular, when there is only one line segment in the selected interval, the inclination angle θ is equal to the angle between the line segment and the x-axis.

[0063] After a series of calculations, the statistical results are as Figure 13As shown in the figure, it is easy to obtain that the interval with the minimum total length of the line segment is [120, 150]. According to the proposed method, the relative optimal inclination angle θ of the scanning line is 135°.

[0064] As Figure 14 shown, internal zigzag trajectories are generated using scanning lines at 45° and 135°. It can be seen from the figure that when the inclination angle is 45°, there are many sharp corners in the internal zigzag trajectory, and when the inclination angle is 135°, there are almost no sharp corners in the internal zigzag trajectory.

[0065] S4. Optimize the torch posture based on collision detection, and adjust the contour trajectory according to the given rules. Specifically: S41. Propose two preconditions to optimize the collision detection algorithm. Refer to Figure 15 , first of all, the torch can be simplified to a cylinder for collision detection with the mold. During the process of repairing the runner crack by WAAR, the robot adopts a simple linear continuous motion mode; and the repaired area after air gouging is in the shape of a "bell mouth", so that even if a collision occurs, the torch posture does not need to be adjusted greatly. Therefore, during the whole repair process, only the collision between the nozzle at the tail of the torch and the repaired area is considered. Secondly, when generating the trajectory, it is necessary to ensure that the distance between adjacent target points in the trajectory is not too far. It can be assumed that when the torch does not collide with the mold at the positions of two target points, the torch will not collide with the mold at any position on the trajectory between the two target points. Therefore, the dynamic collision detection process is transformed into a static collision detection process at each discrete point.

[0066] S42. Collision detection algorithm based on space partitioning. Refer to Figure 16 , in the preliminary detection stage, the mold is divided into a certain number of axis-aligned bounding boxes by space partitioning, and then the collision detection between the bounding boxes and the cylinder is carried out. In the precise detection stage, the Devillers&Guigue algorithm (abbreviated as the Devillers algorithm) is used for the collision detection between the triangular patches and the cylinder. If a collision is detected, the position is recorded as an interference point, and the geometric characteristics (such as angles, distances, etc.) causing the collision are analyzed.

[0067] S43. Adjust the torch posture according to the given rules. The contour trajectory and the zigzag trajectory are adopted in the arc additive manufacturing for repairing the runner crack. The probability of collision inside the zigzag trajectory is relatively low. The present invention only optimizes the posture of the contour trajectory. The rules for adjusting the torch posture of the contour trajectory are as Figure 17 shown. This trajectory is the external contour trajectory, and the point set in it is stored in the counterclockwise direction. Assume that the trajectory is located in a plane parallel to the XY plane, and the torch posture at the current point P1 is adjusted. When the torch in the upright posture collides, the torch posture is rotated clockwise by a certain angle around the forward direction, and the adjustment angle cannot exceed the preset maximum adjustment angle.

[0068] The specific adjustment rules are as follows: Relative to 、 and are the previous point and the next point respectively. The vector is located on the bisector of ∠ The vectors and are located on the central axis of the cylinder, is perpendicular to the plane XY. When the welding torch in the upright position ( ) collides with the mold, it is first adjusted by rotating around in the reference plane . The increment of the adjustment angle is optional, and the adjustment angle cannot exceed the preset maximum adjustment angle. If the welding torch cannot avoid colliding with the mold through adjustment within the plane , then the plane is rotated clockwise or counterclockwise around the vector to obtain a new reference plane, and a new round of detection is performed on it. The increment of the rotation angle of the reference plane is optional, and the rotation angle cannot exceed (half of ∠ ). See Figure 18 for the adjustment result of the contour trajectory.

[0069] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for planning the arc additive repair trajectory of a crack in a hydraulic turbine runner based on an irregular model, characterized in that, It includes the following steps: S1. Use carbon arc air gouging to remove crack defects and establish a model of the blade area to be repaired; S2. Generate a three-dimensional mesh model of the irregular repair area based on reverse reconstruction technology and Boolean operation technology; S3. Optimize the hybrid trajectory planning strategy, slice the repair area to generate the arc additive repair trajectory; S4. Optimize the torch posture based on collision detection and adjust the contour trajectory according to the given rules.

2. The method for arc additive repair trajectory planning of the crack of a hydraulic turbine runner based on an irregular model according to claim 1, wherein, The specific steps in S1 include the following: S11. Detect and evaluate the cracks on the blade surface to clarify the scope and depth of the defects; S12. Use carbon arc air gouging to remove the crack and defective materials. Move the electrode along the crack direction to melt and blow away the materials in the crack and its surrounding fatigue area; S13. Use a grinding wheel or polishing tool to trim the surface after air gouging, remove the oxide layer and slag, and ensure the surface is smooth and free of secondary defects; S14. Use a 3D scanner to accurately scan the blade surface after air gouging to obtain high-resolution point cloud data.

3. The method for planning the arc additive repair trajectory of the crack of the water turbine runner based on the irregular model according to claim 2, wherein, The detection methods for detecting cracks on the blade surface in S11 include: penetrant testing and ultrasonic testing means.

4. The method for planning the arc additive repair trajectory of the crack of the water turbine runner based on an irregular model according to claim 2, wherein, During the accurate scanning of the blade surface after air gouging in S14, ensure that the scanning range includes the complete geometric information of the removal area and its surroundings.

5. The method for planning the arc additive repair trajectory of the crack of the water turbine runner based on the irregular model according to claim 2, wherein, The specific steps in S2 include the following: S21. Perform surface fitting and modeling on the point cloud processed in step S1. Use the least squares method to fit the crack area surface to generate a smooth NURBS surface; convert the fitted surface into a triangular mesh to form a preliminary crack air gouging area model; S22. Use Boolean operation technology to combine the crack area model with the original design model, and generate an accurate three-dimensional model of the repair area through geometric operations; After the three-dimensional mesh model of the repair area is generated through Boolean operation, it needs to be further optimized to ensure its suitability for additive repair trajectory planning; among them, the optimization process includes: removing non-manifold edges, duplicate faces or isolated points generated in Boolean operation; densifying the mesh to accurately describe complex geometric features and ensure that the mesh resolution of the crack repair area meets the requirements of the additive repair trajectory.

6. The method for planning the arc additive repair trajectory of the crack of the water turbine runner based on an irregular model according to claim 2, characterized in that The triangular mesh in S21 adopts the STL format; The specific acquisition process of the accurate three-dimensional model of the repair area in S22 includes: in the CAD modeling software, align the crack area model after air gouging with the complete blade model. First, offset the surface of the designed blade model in the direction of its normal by a certain distance. Subsequently, outline the area to be repaired on the crack area model. Finally, construct the offset surface within the circled area into a solid and perform a Boolean subtraction operation between the solid and the crack area model to obtain the accurate model of the area to be repaired.

7. A method for planning the arc additive repair trajectory of a crack in a water turbine runner based on an irregular model according to claim 2, characterized in that, The specific steps in S3 include the following: S31. Process the unfilled area caused by contour line offset: Calculate the possible unfilled area during the offset process through the contour line offset algorithm. First, offset the model contour line lines1 inward at a given interval to obtain the contour trajectory line lines2; second, offset the obtained contour trajectory line outward by the same distance, denoted as the model contour line lines3; finally, perform a Boolean operation on lines1 and lines3. When the difference set between the two exists and the area is greater than a certain threshold, it is regarded as a valid area for filling. S32. Determine the direction of the scan line: Design a method for selecting the scan line direction to determine the direction of the scan line.

8. A method for planning the arc additive repair trajectory of a crack in a hydraulic turbine runner based on an irregular model according to claim 7, characterized in that The method for selecting the scan line direction in S32 is specifically as follows: The inclination angle θ of the scan line with the positive direction of the X-axis represents the direction of the scan line. First, calculate the length of each line segment in the contour line and its angle with the X-axis, with the angle range being 0° to 180°, evenly divided into 6 intervals; second, calculate the total length of the line segments in each interval; finally, select the interval with the smallest total length of the line segments, and the inclination angle θ is equal to the median value of this interval. Among them, when there is only one line segment in the selected interval, the inclination angle θ is equal to the angle between the line segment and the X-axis.

9. The method for planning the arc additive repair trajectory of the crack of the water turbine runner based on the irregular model according to claim 2, wherein, The specific steps in S4 are as follows: S41. Propose two preconditions to optimize the collision detection algorithm: First, simplify the welding torch into a cylinder for collision detection with the mold. Assume that during the entire repair process, the nozzle at the tail of the welding torch colliding with the mold is the main calculation object. Second, when generating the trajectory, ensure that the distance between adjacent target points in the trajectory is within a certain range; based on the above assumptions, when the welding torch does not collide with the mold at the positions of two target points, the welding torch will not collide with the mold at any position on this trajectory between the two target points. Furthermore, at each discrete point, convert the dynamic collision detection process into a static collision detection process. S42. Preliminary detection and precise detection: In the preliminary detection stage, divide the mold into a certain number of axis-aligned bounding boxes through space partitioning, and then perform collision detection on the bounding boxes and the cylinder; in the precise detection stage, perform collision detection between the triangular patches and the cylinder. S43. The arc additive manufacturing for repairing the runner crack adopts a contour trajectory and a zigzag trajectory.

10. The method for arc additive repair trajectory planning of the crack of a hydraulic turbine runner based on an irregular model according to claim 2, wherein, The specific content of S43 includes: Optimize the posture of the contour trajectory. When the welding torch in the upright posture collides, rotate the posture of the welding torch clockwise by a certain angle around the forward direction, and the adjustment angle cannot exceed the preset maximum adjustment angle; if the welding torch cannot avoid colliding with the repair area, adjust the forward direction and perform a new round of detection; the adjustment amount of the forward direction is optional and limited.

Images