On-machine inspection and compensation method employing point clouds and applied to complex surface processing

Patent metadata

Field	Value
Publication number	US-12061078-B2
Application number	US-201917312543-A
Country	US
Kind code	B2
Filing date	Mar 28, 2019
Priority date	Dec 10, 2018
Publication date	Aug 13, 2024
Grant date	Aug 13, 2024

Abstract

The present application provides an on-machine point cloud detection and compensation method for processing complex surfaces, which comprises: step S 1 , installing a detecting and scanning actuator on an ultrasonic rolling machine tool; step S 2 , installing a processed workpiece on the chuck which is scanned by the detecting and scanning actuator to obtain the point cloud data of the workpiece in a coordinate system of detecting and scanning actuator, which is converted into the point cloud data of the workpiece in a coordinate system of machine tool; step S 3 , processing the point cloud data of the workpiece in the coordinate system of machine tool; step S 4 , obtaining and compensating the shape error feature of the workpiece according to theoretical design data of the processed workpiece and processed point cloud data of the workpiece in the coordinate system of machine tool. The accuracy and efficiency of complex surface strengthening is improved in the present application.

First claim

What is claimed is:

1. An on-machine point cloud detection and compensation method for processing complex surfaces, wherein comprising following steps:
Step S 1 , installing a detecting and scanning actuator on an ultrasonic rolling machine tool, wherein, the ultrasonic rolling machine tool comprises: a chuck and two machining heads respectively installed on left and right sides of the chuck; the detecting and scanning actuator comprises two detecting scanners respectively installed on the machining heads;
Step S 2 , scanning a processed workpiece installed on the chuck by the detecting and scanning actuator to obtain point cloud data of the workpiece in a coordinate system of detecting and scanning actuator, and converting the point cloud data of the workpiece in the coordinate system of detecting and scanning actuator into the point cloud data of the workpiece in a coordinate system of machine tool;
Step S 3 : processing the point cloud data of the workpiece in the coordinate system of machine tool;
Step S 4 : obtaining and compensating shape error feature of the workpiece according to theoretical design data of the processed workpiece and processed point cloud data of the workpiece in the coordinate system of machine tool;
Wherein, the Step S 3 comprises:
Step S 31 , removing external noise points from the point cloud data of the workpiece in the coordinate system of machine tool by means of a main point cloud cluster extraction method, so as to obtain an original point cloud data of a blade model;
Step S 32 , removing original data noise points from the original point cloud data of the blade model by means of statistical outlier method so as to obtain denoised point cloud data;
Step S 33 , simplifying and smoothing the denoised point cloud data by means of bounding box method considering curvature.

2. The on-machine point cloud detection and compensation method for processing complex surfaces according to claim 1 , the Step S 31 comprises:
clustering the point cloud data of the workpiece in the coordinate system of machine tool according to density, wherein point cloud with highest density is main point cloud data of blade, which is recorded as MP, point cloud with low density is recorded as OP i , distance from OP i to MP is calculated according to following equation, if the distance exceeds a predetermined thresholdD τ , it means OP i is the external noise point to be deleted;
Dist( OP i ,MP )=∥ OP i −MP∥   (2).

3. The on-machine point cloud detection and compensation method for processing complex surfaces according to claim 1 , wherein the Step S 32 comprises:
calculating average distance d mid (P i ) of distances from each point P i of the original point cloud of the blade model to all points of neighborhood M according to following equation, if d mid (P i )>D mid , point P i is defined as an outlier, and finally the outlier is deleted, wherein D mid is global average distance of model;

d
mid

⁡

(

P
i

)

=

1
k

⁢

∑

j
=
1

k

⁢



P
i

-

M
j



,

wherein k is number of point cloud in neighborhood M of point P i , M j is point in neighborhood M.

4. The on-machine point cloud detection and compensation method for processing complex surfaces according to claim 1 , wherein the Step S 33 comprises: putting the denoised point cloud data into a cuboid, dividing the cuboid evenly into cubes with equal edges according to a predetermined simplification rate; for each cube, selecting the point cloud data closest to center of the cube as feature point, and reducing edge length of the cube to increase its density for areas with curvature greater than 0.33 in the denoised point cloud data.

5. The on-machine point cloud detection and compensation method for processing complex surfaces according to claim 4 , wherein the Step S 33 also comprises: calculating curvature of the denoised point cloud data by means of conicoid fitting method.

6. The on-machine point cloud detection and compensation method for processing complex surfaces according to claim 1 , wherein the shape error feature of the workpiece comprises: maximum deviation values in X and Y directions between actual position data on X axis and Y axis of each layer of section of the processed workpiece and design position data on X axis and Y axis of each layer of section of the processed workpiece, the Step S 4 comprises: compensating the maximum deviation value in X direction by means of force feedback compensation method, and compensating the maximum deviation value in Y direction by means of increasing or decreasing trajectory offset of the machining heads in Y direction.

7. The on-machine point cloud detection and compensation method for processing complex surfaces according to claim 1 , wherein the ultrasonic rolling machine tool also comprises: a bed base, the chuck can rotate around Z axis and be movably mounted on the bed base along Z axis.

8. The on-machine point cloud detection and compensation method for processing complex surfaces according to claim 1 , wherein the detecting and scanning actuator also comprises: two Y axis lifting mobile worktables installed on the bed base and symmetrically distributed on left and right sides of the processed workpiece, and two X axis left and right mobile worktables respectively installed on the Y axis lifting mobile worktables.

9. The on-machine point cloud detection and compensation method for processing complex surfaces according to claim 1 , wherein the detecting scanner is a laser scanner.

Assignees

• Univ East China Science & Tech

Inventors

• Zhang Xiancheng
• Liu Shuang
• Yao Shulei
• Zhu Xumin
• Zhang Kaiming
• Liu Yixin
• Jia Yunfei
• Tu Shantung

Classifications

• G01B11/002

  for measuring two or more coordinates · CPC title

• B23Q17/2457

  of tools · CPC title

• B23Q17/098

  by measuring noise · CPC title

• G01B11/245

  using a plurality of fixed, simultaneously operating transducers ({G01B11/2408 - G01B11/2425, } G01B11/255 take precedence) · CPC title

• G01B11/25

  by projecting a pattern, e.g. {one or more lines,} moiré fringes on the object (G01B11/255 takes precedence {; image analysis for depth or shape recovery G06T7/50}) · CPC title