Optimization method of laser scanning trajectory for film holes of turbine blades based on collision detection

Abstract

A method for optimizing the laser scanning trajectory of the air film hole of a turbine blade based on collision detection, firstly, the three-dimensional model grid of the laser sensor is numerically processed, and after coordinate transformation, a point cloud scanning volume of the sensor moving according to the initial planning trajectory is generated; then, the bounding box of the area to which the turbine blade and the fixture to be measured is obtained according to the geometric characteristics and structural characteristics, and the hardware collision detection is performed through the separation axis projection of the point cloud scanning volume and the bounding box to obtain the collision information; finally, according to the collision information, the obstacle avoidance statement is embedded in the initial planning trajectory and a collision-free optimization trajectory is generated. The present invention effectively avoids complex spatial geometric operations and is easy to implement in programming.

Description

Technical Field

[0001] The invention relates to a technology in the field of aeroengine manufacturing, and in particular to a collision detection-based method for optimizing the laser scanning trajectory of a turbine blade air film hole of an aeroengine. Background Art

[0002] Turbine blades are one of the key components of aircraft engines, and the film cooling holes on them are crucial to the cooling performance of the engine. In order to ensure that the geometric parameters of such cooling holes, such as position, axis direction, and aperture size, meet the design requirements and have sufficient reliability, their geometric characteristics need to be tested after the holes are made.

[0003] Existing detection methods mostly use laser sensors to scan the blades to generate point clouds, and then analyze the point clouds to obtain the geometric parameters of the film cooling holes. However, the surface contour of the turbine blade is a complex free-form surface, and the number of film cooling holes is large and the distribution is complex. To achieve complete and comprehensive scanning and detection, it is necessary to pre-plan the movement path of the laser sensor. Considering the actual production needs, the planned trajectory must ensure safety while meeting the requirements of efficient and accurate detection. That is, the laser sensor must not collide with hardware equipment such as turbine blades, fixtures, and machine tool bodies during trajectory movement. Therefore, collision detection must be performed during trajectory planning, and detected collisions must be avoided. Summary of the invention

[0004] In view of the defects and shortcomings of the prior art, the present invention proposes a method for optimizing the laser scanning trajectory of the air film holes of turbine blades based on collision detection, which effectively avoids complex spatial geometric calculations and is easy to implement through programming.

[0005] The present invention is achieved through the following technical solutions:

[0006] The present invention relates to a method for optimizing the laser scanning trajectory of an air film hole of a turbine blade based on collision detection. Firstly, the three-dimensional model grid of a laser sensor is numerically processed, and a point cloud scanning body of the sensor moving according to an initial planned trajectory is generated after coordinate transformation; then, a bounding box of the area to which the turbine blade and a fixture to be measured are obtained according to their geometric features and structural characteristics, and hardware collision detection is performed through the separation axis projection of the point cloud scanning body and the bounding box to obtain collision information; finally, an obstacle avoidance statement is embedded in the initial planned trajectory according to the collision information, and a collision-free optimized trajectory is generated.

[0007] The coordinate transformation is performed with the help of the initial planning trajectory and the DH matrix. First, the three-dimensional model of the laser sensor is grid-digitized to generate a point cloud Pc l ; Then the G code of the initial trajectory is processed into incremental mode and subdivided to obtain the subdivided incremental G code C sub, and then C sub The coordinate increment in X i , Y i , Z i , A i , B i Based on the transformation matrix T i Point Cloud Perform spatial transformation processing and compare the obtained point cloud with the previous result Overlay to get the scan volume of the current laser sensor point cloud

[0008] The transformation matrix T i =G T (X i ,Y i ,Z i )×Rot(x,A i )×Rot(y,B i ), where: translation transformation matrix A-axis rotation matrix B-axis rotation matrix

[0009] The bounding box is the smallest regular geometric body B in the space that can enclose the turbine blade to be tested and the area where the fixture moves. v , which can be a cylinder or a cuboid, is determined by the geometrical and structural features of the turbine blade to be measured and the fixture itself, combined with the area to which it moves.

[0010] The hardware collision detection refers to: the laser sensor and the three-dimensional features of the turbine blade and the fixture to be tested are separated by axis projection, that is, the scanning volume under the same coordinate reference is and bounding box B v Project onto the X, Y, and Z coordinate axes. When the projections in each axis overlap, a collision is determined and the corresponding collision information is recorded.

[0011] The collision information refers to relevant parameters obtained through collision detection that can be used for subsequent collision avoidance, including obstacle avoidance time, obstacle avoidance direction and obstacle avoidance distance.

[0012] The obstacle avoidance moment, i.e., the G code sequence node for the start and end of collision avoidance, includes the start time t s and the end time t e , when scanning the body With bounding box B v When a collision occurs, record t s =i, and let i increase by 1, perform separation axis projection collision detection, when i accumulates to i', no collision occurs, then record t e =i'.

[0013] The obstacle avoidance direction is the direction in which the laser sensor moves to avoid collision. a , which is determined based on the mutual motion relationship between the sensor and the workpiece.

[0014] The obstacle avoidance distance is the distance that the laser sensor moves along the obstacle avoidance direction v when performing collision avoidance. a The minimum distance S required to move h =K s (s max -s min ), where: K s is the collision avoidance safety factor; s max 、s min It means: will be at time t s and t e The point cloud of the collision area between them moves towards the obstacle avoidance direction v a The projection contour is at v a Maximum and minimum values ​​in a direction.

[0015] The collision-free optimized trajectory refers to: according to the obstacle avoidance time, obstacle avoidance direction and obstacle avoidance distance in the collision information, the initial G code C sub Insert obstacle avoidance statements in the code to make the newly generated trajectory avoid the collision area while keeping other parts unchanged. Specifically:

[0016] ① At the obstacle avoidance start time t s Forward, toward C sub Insert the obstacle avoidance start code in:

[0017] ② At the end of obstacle avoidance time t e Afterwards, to C sub Insert the obstacle avoidance end code in: in: Obstacle avoidance distances on the X, Y, and Z axes respectively in: Represents vector v a Projection onto each axis.

[0018] Technical Effects

[0019] The present invention solves the problem that the laser sensor scanning track cannot perform rapid collision detection and collision avoidance during the existing turbine blade film cooling hole detection process.

[0020] Compared with the prior art, the present invention performs numerical grid processing on the three-dimensional model and performs collision judgment through coordinate projection results, so as to obtain quantitative information such as obstacle avoidance time, obstacle avoidance direction, obstacle avoidance distance, etc., avoiding complex spatial geometric calculations. It is only necessary to insert obstacle avoidance start and end statements in the initial trajectory to generate a new collision-free trajectory, which completes the collision avoidance without affecting the normal non-collision part of the trajectory. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 is a flow chart of the collision detection method;

[0022] Figure 2 It is a flow chart of the collision avoidance method;

[0023] Figure 3 It is a schematic diagram of the turbine blade to be tested and the bounding box of the fixture;

[0024] Figure 4 It is a schematic diagram of the laser sensor scanning body;

[0025] Figure 5 It is a schematic diagram of the projection of the scanned body and the bounding box on each coordinate plane when a collision occurs;

[0026] Figure 6 Schematic diagram of the collision avoidance process. DETAILED DESCRIPTION

[0027] The present embodiment relates to a turbine blade air film hole laser scanning trajectory optimization system based on collision detection, comprising: a collision detection module and a trajectory optimization module, wherein: the collision detection module performs hardware collision detection and outputs collision information, and the trajectory optimization module performs collision avoidance according to the collision information and generates an optimized collision-free scanning trajectory.

[0028] like Figure 1 and Figure 2 As shown, the laser scanning trajectory optimization method of the turbine blade film hole of the above system is shown, and the specific steps include:

[0029] Step 1) Process the G code of the initial trajectory into incremental mode and subdivide it to obtain the subdivided incremental G code C sub .

[0030] Step 2) Based on the geometric parameters of the turbine blade and the fixture to be tested and the motion trajectory during the detection process, generate the minimum cylinder that can envelop the area where the turbine blade and the fixture to be tested move, i.e., the bounding box B v ,like Figure 3 shown.

[0031] Step 3) Grid the 3D model of the laser sensor to generate a point cloud Pc l .

[0032] Step 4) C sub The coordinate increment in X i , Y i , Z i , A i , B i Based on the transformation matrix T i Point Cloud PC l Perform spatial transformation processing and compare the obtained point cloud with the previous result Overlay to get the scan volume of the current laser sensor point cloud like Figure 4 shown.

[0033] The transformation matrix T i =G T (X i ,Y i ,Z i )×Rot(x,A i )×Rot(y,B i ), where: translation transformation matrix A-axis rotation matrix B-axis rotation matrix

[0034] Step 5) Scan the volume under the same coordinate reference and bounding box B v Projection to the X, Y, and Z coordinate axes. Figure 5 As shown, if the projections in each axis overlap, a collision occurs; otherwise, no collision occurs.

[0035] Step 6) When step (5) detects that a collision occurs, three parameters, namely, obstacle avoidance time, obstacle avoidance direction, and obstacle avoidance distance, are determined.

[0036] The obstacle avoidance time refers to the G code sequence nodes of the collision avoidance start and end, including the start time t s and the end time t e When scanning the body With bounding box B v When a collision occurs, record t s =i, and let i be incremented by 1, and execute the separation axis projection detection described in step (5). If no collision occurs when i is accumulated to i', then record t e =i'.

[0037] The obstacle avoidance direction refers to the direction in which the laser sensor moves to avoid collision. Combined with the basic form of mutual movement between the sensor and the workpiece, the obstacle avoidance direction is taken as the Z-axis direction.

[0038] The obstacle avoidance distance is the minimum distance S that the laser sensor needs to move along the obstacle avoidance direction when performing collision avoidance. h , the calculation formula is: S h =K s (z max -z min ), where: K s is the collision avoidance safety factor; z max 、z min The calculation method is: at time t s and t e The point cloud of the collision area between the two is projected onto the Z axis where the obstacle avoidance direction is located. The maximum and minimum values ​​of the projection contour are z max 、z min .

[0039] Step 7) Based on the obstacle avoidance parameters determined in step (6), in the initial G code C sub The obstacle avoidance statement is inserted into the new trajectory so that the newly generated trajectory avoids the collision area while the other parts remain unchanged.

[0040] The method of inserting the obstacle avoidance statement is as follows: at the obstacle avoidance start time t s Forward, toward C sub Insert the obstacle avoidance start code in:

[0041] C StartAvoid :X0 Y0 ZS h B0C0

[0042] In C sub The obstacle avoidance end time t e Finally, insert the obstacle avoidance end code:

[0043] C EndAvoid :X0 Y0 Z(-S h ) B0 C0

[0044] The process of inserting obstacle avoidance statements to avoid the trajectory is shown as follows: Figure 6 shown.

[0045] This can complete the collision detection and avoidance of the laser sensor scanning trajectory and generate a new collision-free motion trajectory.

[0046] In the above operation, the laser sensor and the turbine blades and fixture to be measured are grid-digitized to extract key features, and the initial motion trajectory is incrementally subdivided and obstacle avoidance start and end statements are inserted to avoid collision, which is original to the present invention.

[0047] The grid numerical processing is specifically as follows: the three-dimensional solid model of the laser sensor is sparsely processed to obtain the initial point cloud of the scanned body; based on the geometric features and structural characteristics of the turbine blades and fixtures to be measured and the area to which they move, they are abstracted into a bounding box composed of discrete points whose outer envelope is a regular geometric body.

[0048] The insertion of obstacle avoidance start and end statements for collision avoidance is specifically as follows: based on the collision information obtained by collision detection, an obstacle avoidance start code and an obstacle avoidance end code are generated, and are inserted into the subdivided incremental G code before the obstacle avoidance start time and after the obstacle avoidance end time respectively, while keeping the other parts of the G code unchanged.

[0049] The method proposed in the present invention was verified by using VERICUT CNC machining simulation software. Fifteen groups of collision trajectories at different positions and times were designed. The algorithm of the present invention was called to perform collision detection and avoidance. The newly generated trajectories were simulated by VERICUT software. The results showed that the collision-free trajectories accounted for 100%, and the new trajectories maintained the basic characteristics of the original trajectories.

[0050] Compared with the existing technology, this method effectively avoids complex spatial geometric operations and is easy to implement through programming; the collision detection algorithm is stable and reliable, and the optimization process has little impact on the original trajectory, which is suitable for industrial applications.

[0051] The above-mentioned specific implementation can be partially adjusted in different ways by those skilled in the art without departing from the principle and purpose of the present invention. The protection scope of the present invention shall be based on the claims and shall not be limited by the above-mentioned specific implementation. Each implementation scheme within its scope shall be subject to the constraints of the present invention.

Claims

1. A method for optimizing the laser scanning trajectory of a turbine blade film hole based on collision detection, characterized in that: Firstly, the 3D model grid of the laser sensor is numerically processed, and after coordinate transformation, a point cloud scan volume of the sensor moving according to the initial planned trajectory is generated; then, the bounding box of the area to which the turbine blade and the fixture to be measured are obtained according to the geometric features and structural characteristics, and hardware collision detection is performed through the separation axis projection of the point cloud scan volume and the bounding box to obtain collision information; finally, according to the collision information, the obstacle avoidance statement is embedded in the initial planned trajectory and a collision-free optimized trajectory is generated; The collision information includes: obstacle avoidance time, obstacle avoidance direction and obstacle avoidance distance; The bounding box is the smallest regular geometric body B in the space that can enclose the turbine blade to be tested and the area where the fixture moves. v , which can be a cylinder or a cuboid, is determined by the geometrical and structural features of the turbine blade to be tested and the fixture itself, and in combination with the area to which it moves; The obstacle avoidance moment, i.e., the G code sequence node for the start and end of collision avoidance, includes the start time t s and the end time t e , when scanning the body With bounding box B v When a collision occurs, record t s =i, and let i increase by 1, perform separation axis projection collision detection, when i accumulates to i′, no collision occurs, then record t e =i′; The obstacle avoidance direction is the direction in which the laser sensor moves to avoid collision. a , which is determined according to the mutual motion relationship between the sensor and the workpiece; The obstacle avoidance distance is the distance that the laser sensor moves along the obstacle avoidance direction v when performing collision avoidance. a The minimum distance S required to move h =K s (s max -s min ), where: K s is the collision avoidance safety factor; s max 、s min It means: will be at time t s and t e The point cloud of the collision area between them moves towards the obstacle avoidance direction v a The projection contour is at v a Maximum and minimum values ​​in direction; The collision-free optimized trajectory refers to: according to the obstacle avoidance time, obstacle avoidance direction and obstacle avoidance distance in the collision information, the initial G code C sub Insert obstacle avoidance statements in the code to make the newly generated trajectory avoid the collision area while keeping other parts unchanged. Specifically: ① At the obstacle avoidance start time t s Forward, toward C sub Insert the obstacle avoidance start code in: ② At the end of obstacle avoidance time t e Afterwards, to C sub Insert the obstacle avoidance end code in: in: Obstacle avoidance distances on the X, Y, and Z axes respectively in: Represents vector v a Projection onto each axis.

2. The method for optimizing the laser scanning trajectory of the air film hole of a turbine blade based on collision detection according to claim 1 is characterized in that: The coordinate transformation is performed with the help of the initial planning trajectory and the DH matrix. First, the three-dimensional model of the laser sensor is grid-digitized to generate a point cloud Pc l ; Then the G code of the initial trajectory is processed into incremental mode and subdivided to obtain the subdivided incremental G code C sub , and then C sub The coordinate increment in X i , Y i , Z i , A i , B i Based on the transformation matrix T i Point Cloud PC l Perform spatial transformation processing and compare the obtained point cloud with the previous result Overlay to get the scan volume of the current laser sensor point cloud 3. The method for optimizing the laser scanning trajectory of the air film hole of a turbine blade based on collision detection according to claim 2 is characterized in that: The transformation matrix T i =G T (X i ,Y i ,Z i )×Rot(x,A i )×Rot(y,B i ), where: translation transformation matrix A-axis rotation matrix B-axis rotation matrix 4. The method for optimizing the laser scanning trajectory of the air film hole of a turbine blade based on collision detection according to claim 1 is characterized in that: The hardware collision detection refers to: the laser sensor and the three-dimensional features of the turbine blade and the fixture to be tested are separated by axis projection, that is, the scanning volume under the same coordinate reference is and bounding box B v Project onto the X, Y, and Z coordinate axes. When the projections in each axis overlap, a collision is determined and the corresponding collision information is recorded.

5. A system for optimizing the laser scanning trajectory of a turbine blade film hole based on collision detection, which implements the method described in any one of claims 1 to 4, characterized in that: include: A collision detection module and a trajectory optimization module, wherein: the collision detection module performs hardware collision detection and outputs collision information, and the trajectory optimization module performs collision avoidance according to the collision information and generates an optimized collision-free scanning trajectory.

Images