WO2020118970A1 - 一种复杂表面加工的在机点云检测及补偿方法 - Google Patents

一种复杂表面加工的在机点云检测及补偿方法 Download PDF

Info

Publication number
WO2020118970A1
WO2020118970A1 PCT/CN2019/080138 CN2019080138W WO2020118970A1 WO 2020118970 A1 WO2020118970 A1 WO 2020118970A1 CN 2019080138 W CN2019080138 W CN 2019080138W WO 2020118970 A1 WO2020118970 A1 WO 2020118970A1
Authority
WO
WIPO (PCT)
Prior art keywords
point cloud
workpiece
cloud data
machine
detection
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2019/080138
Other languages
English (en)
French (fr)
Inventor
张显程
刘爽
姚树磊
朱旭旻
张开明
刘怡心
贾云飞
涂善东
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
East China University of Science and Technology
Original Assignee
East China University of Science and Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by East China University of Science and Technology filed Critical East China University of Science and Technology
Priority to US17/312,543 priority Critical patent/US12061078B2/en
Publication of WO2020118970A1 publication Critical patent/WO2020118970A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00—Measuring arrangements characterised by the use of optical techniques
    • G01B11/24—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
    • B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
    • B23Q17/00—Arrangements for observing, indicating or measuring on machine tools
    • B23Q17/24—Arrangements for observing, indicating or measuring on machine tools using optics or electromagnetic waves
    • B23Q17/2452—Arrangements for observing, indicating or measuring on machine tools using optics or electromagnetic waves for measuring features or for detecting a condition of machine parts, tools or workpieces
    • B23Q17/2471—Arrangements for observing, indicating or measuring on machine tools using optics or electromagnetic waves for measuring features or for detecting a condition of machine parts, tools or workpieces of workpieces
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
    • B23P—METAL-WORKING NOT OTHERWISE PROVIDED FOR; COMBINED OPERATIONS; UNIVERSAL MACHINE TOOLS
    • B23P9/00—Treating or finishing surfaces mechanically, with or without calibrating, primarily to resist wear or impact, e.g. smoothing or roughening turbine blades or bearings; Features of such surfaces not otherwise provided for, their treatment being unspecified
    • B23P9/04—Treating or finishing by hammering or applying repeated pressure
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
    • B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
    • B23Q17/00—Arrangements for observing, indicating or measuring on machine tools
    • B23Q17/09—Arrangements for observing, indicating or measuring on machine tools for indicating or measuring cutting pressure or for determining cutting-tool condition, e.g. cutting ability, load on tool
    • B23Q17/0952—Arrangements for observing, indicating or measuring on machine tools for indicating or measuring cutting pressure or for determining cutting-tool condition, e.g. cutting ability, load on tool during machining
    • B23Q17/098—Arrangements for observing, indicating or measuring on machine tools for indicating or measuring cutting pressure or for determining cutting-tool condition, e.g. cutting ability, load on tool during machining by measuring noise
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
    • B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
    • B23Q17/00—Arrangements for observing, indicating or measuring on machine tools
    • B23Q17/24—Arrangements for observing, indicating or measuring on machine tools using optics or electromagnetic waves
    • B23Q17/2452—Arrangements for observing, indicating or measuring on machine tools using optics or electromagnetic waves for measuring features or for detecting a condition of machine parts, tools or workpieces
    • B23Q17/2457—Arrangements for observing, indicating or measuring on machine tools using optics or electromagnetic waves for measuring features or for detecting a condition of machine parts, tools or workpieces of tools
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00—Measuring arrangements characterised by the use of optical techniques
    • G01B11/002—Measuring arrangements characterised by the use of optical techniques for measuring two or more coordinates
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00—Measuring arrangements characterised by the use of optical techniques
    • G01B11/24—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures
    • G01B11/245—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures using a plurality of fixed, simultaneously operating transducers
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00—Measuring arrangements characterised by the use of optical techniques
    • G01B11/24—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures
    • G01B11/25—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures by projecting a pattern, e.g. one or more lines, moiré fringes on the object

Definitions

  • the invention relates to an on-machine point cloud detection and compensation method for complex surface processing.
  • the existing measurement methods for complex curved workpieces mainly include: standard sample method, automatic drawing measurement method, optical projection measurement method, coordinate measurement method and inductance measurement method, among which: standard sample method is a qualitative
  • standard sample method is a qualitative
  • the measurement method can only be used to judge whether the part is qualified or not, and the actual error of the workpiece cannot be given; although the measurement accuracy of the automatic drawing measurement method is higher than the standard model method, it still cannot meet the needs of high-precision blade detection, and automatic drawing
  • the measuring instrument is bulky and inconvenient to operate, so its application range is limited; the optical projection measurement method is generally used for qualitative detection. By observing the degree of conformity between the actual leaf shape and the theoretical leaf shape, the blade is directly qualified, but it is quantified by reading the coordinate value.
  • the efficiency is relatively low, and it is limited by its own display screen, the measurement range is limited, and at the same time it is greatly affected by the reflection ability of the workpiece surface, and the measurement accuracy is low;
  • the coordinate measurement method is a general point measurement equipment, which has high cost, The measurement process is cumbersome, the measurement efficiency is low, the secondary development of measurement software is required, and the requirements on the environment are high.
  • the inductance measurement method is not complete contour detection, so the measurement results are not very reliable, there is a possibility of misjudgment, and the need for different workpieces Equipped with corresponding bracket devices and standard workpieces, increasing the measurement cost and time.
  • the development of optical measurement technology has provided a new measurement method for complex curved surface measurement.
  • People can use advanced complex curved surface shape detection equipment, such as metal surface defect detectors; 3D scanners (including laser 3D scanners, The structured light 3D scanner realizes the accurate detection and error evaluation of the workpiece surface.
  • the new measurement methods mainly include: machine vision measurement method, four-coordinate laser measurement method and laser scanning measurement method, etc. These measurements The advantages of the method are fast measurement speed, large range and high efficiency.
  • the various probes and corresponding software installed in the machining center specially developed by Reni Shaw of the United Kingdom can realize functions such as tool setting, tool breakage detection, and workpiece error detection; on-machine gear measurement device of German Kapp gear grinding machine ;
  • the well-known gear machine tool manufacturers Gleason and Klinger Bell connect the gear testing center with the processing machine tool to realize the closed-loop manufacturing of CAD/CAM/CAI of the gear;
  • the general inspection system Power of British Delcam Company INSPECT can Realize on-line detection and offline detection.
  • domestic scholars have also carried out exploration and research work on machine detection technology. Related research mainly includes: comprehensive error modeling of machine tools, measurement and identification of error parameters, automatic planning of workpiece detection paths, and compensation techniques for different error sources.
  • existing error compensation methods such as multiple compensation methods based on inverse deformation errors, have the disadvantages of low error compensation efficiency.
  • the contact-type detection probe directly contacts the object to be measured, triggering the probe sensor to send a measurement signal, and receiving the sensor to receive the signal , Determine the current spherical center coordinates of the probe, and finally calculate the coordinate value of the point through probe radius compensation and error compensation. It has the characteristics of high accuracy and strong environmental adaptability, but it has the disadvantages of low detection efficiency and high detection cost; Non-contact detection probes use optical and other physical characteristics to reverse the shape information of objects. Compared with contact detection methods, they have the characteristics of fast scanning speed, can obtain the overall contour of the measured workpiece, and can measure the surface.
  • the soft workpiece has high precision, which can reach an accuracy of 0.01mm, but it is also easily affected by the reflection characteristics of the workpiece surface (such as curvature, color, roughness, etc.).
  • measurement point cloud For freeform surfaces such as thin-walled aerospace blades, it is often necessary to repeatedly measure multiple times, resulting in a measurement point cloud scale of 100,000/million; more seriously, it is affected by factors such as temperature differences, dust particles, and small vibrations in the industrial site.
  • the measurement point cloud has inherent defects (residual noise, extracorporeal holes, stacked areas, density differences, etc.).
  • the resulting large-scale measurement point cloud data processing has become one of the key problems in the detection of complex surfaces, such as streamlining, smoothing, flattening, and reconstruction geometric operations; specifically, due to the insufficient wall thickness of the front and rear edges of the aviation blade 2 mm, the front and rear edge data are easily lost during the conventional streamlining or smoothing process; due to the non-uniform distribution of the blade point cloud itself, the sparse point cloud should be properly retained in the sparse point cloud distribution area during the streamlining process, otherwise it will cause voids; in addition, Avoid losing the high curvature characteristics of the blade edge due to excessive smoothing.
  • the present invention aims to provide an on-machine point cloud detection and compensation method for complex surface processing, so as to efficiently and quickly detect and compensate errors in the process of ultrasonic rolling strengthening processing.
  • the on-machine point cloud detection and compensation method for complex surface processing according to the present invention includes the following steps:
  • Step S1 Install an inspection scanning execution mechanism on the ultrasonic rolling machine, wherein the ultrasonic rolling machine includes: a chuck and two processing heads respectively installed on the left and right sides of the chuck; the inspection scan The execution mechanism includes: two detection scanners respectively installed on the processing head;
  • Step S2 Scan the workpiece mounted on the chuck by using the scanning execution mechanism to obtain the workpiece point cloud data in the coordinate system of the inspection equipment, and the workpiece point cloud data in the coordinate system of the inspection equipment Converted to the workpiece point cloud data in the machine tool coordinate system;
  • Step S3 Perform data processing on the workpiece point cloud data in the machine tool coordinate system
  • Step S4 Obtain the workpiece shape error feature based on the theoretical design data of the workpiece to be processed and the workpiece point cloud data of the machine tool coordinate system after data processing, and compensate the workpiece shape error feature;
  • step S3 includes:
  • Step S31 the subject point cloud cluster extraction method is used to remove the extracorporeal noise points in the workpiece point cloud data under the machine tool coordinate system to obtain the original point cloud data of the blade model;
  • Step S32 using the statistical outlier method to remove the original data noise points in the original point cloud data of the blade model to obtain noise-reduced point cloud data;
  • step S33 the bounding box method considering curvature is used to simplify and smooth the noise-reduced point cloud data.
  • the step S31 includes:
  • the workpiece point cloud data in the machine tool coordinate system is clustered according to the density.
  • the point cloud set with the highest density is the main point cloud data of the blade, which is denoted as MP, and the point cloud set with the lower density is denoted as OP i , according to the following formula Calculate the distance between OP i and MP. If the distance exceeds a set threshold D ⁇ , it means that OP i is the extracorporeal noise point to be deleted;
  • the step S32 includes:
  • k represents the number of point clouds in the neighborhood M of the point P i and M j represents the j-th point in the neighborhood M.
  • the step S33 includes: putting the noise-reduced point cloud data into a rectangular parallelepiped, and according to the size of the preset reduction rate
  • the cuboid is evenly divided into equilateral cubes. For each cube, the point cloud data closest to the center of the cube is selected as a feature point, and for the area of the denoised point cloud data with a curvature greater than 0.33, the cube is reduced. Side length to increase its density.
  • the step S33 further includes: using a quadratic surface fitting method to calculate the curvature of the noise-reduced point cloud data.
  • the workpiece shape error features include: the actual position data of each section of the processed workpiece on the X-axis and Y-axis and the processed The maximum deviation value in the X direction and the Y direction between the design position data on the X axis and the Y axis of each section of the workpiece, the step S4 includes: using the force feedback compensation method The maximum deviation value is compensated, and the maximum deviation value in the Y direction is compensated by increasing or decreasing the trajectory offset of the processing head in the Y direction.
  • the ultrasonic rolling machine tool further includes: a bed base, and the chuck can rotate around the Z axis and be movably installed along the Z axis On the bed base.
  • the detection scan execution mechanism further includes: two Y mounted on the bed base and symmetrically distributed on the left and right sides of the workpiece An axis moving table and two X-axis moving tables respectively mounted on the Y-axis moving table.
  • the detection scanner is a laser scanner.
  • the present invention uses non-contact detection technology to perform the work of the workpiece during the surface strengthening process of the complex curved workpieces
  • On-machine inspection that is, scanning the surface of the workpiece surface at the processing station of the complex surface workpiece, quickly obtaining the shape data of the complex surface, and obtaining the comprehensive error of the shape position of the complex surface through large-scale point cloud processing technology, and analyzing the point cloud
  • the characteristic of the difference between the data model and the design model is to obtain an error compensation scheme based on force feedback and position compensator (that is, to output compensation parameters to the position compensator of the machine tool subband) to perform trajectory error compensation, thereby improving the surface of complex curved surfaces Strengthen the accuracy and efficiency of processing.
  • FIG. 1a is a schematic structural view of the detection scanning actuator installed on an ultrasonic rolling machine in a method for detecting and compensating a complex surface of the present invention
  • 1b is a schematic diagram of the structure of the workpiece, chuck, processing head and detection scanner in the present invention, which shows the X, Y, Z axis movement direction of the chuck and detection scanner;
  • FIG. 2 is a schematic diagram of a coordinate system of a detection device and a coordinate system of a machine tool in the present invention
  • FIG. 3 is a schematic diagram of the original point cloud data in the present invention.
  • step S3 is a schematic diagram of point cloud data processed in step S3 in the present invention
  • FIG. 5 is a schematic diagram of error characteristic analysis on the X-Y plane of a certain processing section of the blade in the present invention.
  • FIG. 6 is a schematic diagram of using the force feedback compensation method to compensate errors in the X direction in the present invention.
  • the invention that is, an on-machine point cloud detection and compensation method for complex surface processing, includes the following steps:
  • Step S1 Install an inspection and scanning execution mechanism on the ultrasonic rolling machine, wherein,
  • the ultrasonic rolling machine includes: a chuck 3 installed on the bed base 1 and for mounting the workpiece 2 to be processed thereon, wherein the chuck 3 can rotate around the Z axis and can be placed on the bed
  • the body base 1 moves along the Z axis, and the end of the processed workpiece 2 is installed on one end surface of the chuck 3, that is, the chuck 3 can drive the processed workpiece 1 to rotate around the Z axis and feed in the Z axis direction;
  • the scanning execution mechanism includes: two Y-axis lifting and moving worktables 4 installed on the bed base 1 and symmetrically distributed on the left and right sides of the workpiece 2 to be processed, and two X-axis mounted on the Y-axis lifting and moving worktables 4, respectively Left and right moving table 5, two processing heads 6 respectively mounted on the X-axis moving table 5 (processing head 6 is an inherent part of the ultrasonic rolling machine), and two detection scanners 7 respectively mounted on the processing head 6 That is, the Y-axis up and down moving table 4 and the X-axis moving table 5 can drive the detection scanner 7 to move up and down in the Y-axis direction and to move left and right in the X-axis direction;
  • the workpiece 1 to be processed is an aircraft engine blade (there are many types of engine blades, and their blade shapes are mostly free-form curved surfaces; the invention is applicable to double-arc, multi-arc or S-blade type blades, and The maximum blade length is 1200mm); due to the problem of the reflection of the blade surface, through the analysis and comparison of the point cloud effect generated by the laser scanner and the structured light scanner on the scanning of complex curved workpieces, the comprehensive scanner cost, size and scanner requirements are finalized.
  • the laser scanner is selected as the detection scanner 7 for blade point cloud acquisition. Specifically, the structured light scanner extracts three-dimensional data on the surface of the object by taking pictures, and the laser scanner extracts three-dimensional data on the surface of the object through line scanning.
  • the metallic surface is shiny and reflective, which will seriously affect the effect of structured light scanners to obtain point clouds, and laser scanners can avoid this problem, and the point clouds obtained by laser scanners
  • the accuracy of the point cloud obtained by the structured light scanner is higher, that is, the quality of the point cloud data is higher.
  • the laser scanner is more portable than the structured light scanner, and it can perform any angle-free dead angle scanning at the operation site; in this implementation
  • the size of the laser scanner is about 20mm*30mm*10, and its working characteristics are: fixed automatic scanning, high efficiency, high accuracy, and high resolution.
  • Step S2 using a scanning actuator, that is, using the detection scanner to move up and down in the Y-axis direction and left and right in the X-axis direction, combined with the feed of the chuck 3 in the Z-axis direction and rotation around the Z-axis, Scan the processed workpiece 2 to obtain the workpiece point cloud data in the coordinate system of the detection device, and convert the workpiece point cloud data in the coordinate system of the detection device into the workpiece point cloud data in the machine tool coordinate system;
  • a scanning actuator that is, using the detection scanner to move up and down in the Y-axis direction and left and right in the X-axis direction, combined with the feed of the chuck 3 in the Z-axis direction and rotation around the Z-axis
  • the coordinate system of the testing equipment is different from the coordinate system of the machine tool, the coordinate system needs to be converted and parameter calibration is performed (parameter calibration is to locate the relative coordinate of the machine tool absolute coordinate system after installing the detection mechanism and clamping the workpiece to be processed each time , The calibrated rotation angle information will ensure that it is at the true zero position at the beginning of the detection).
  • the coordinate system of the detection device can be converted into the coordinate system of the machine tool:
  • P i represents the machine tool coordinate system
  • P j represents the detection equipment coordinate system
  • I represents the rotation matrix, which represents the amount of rotation of the detection device coordinate system relative to the initial position
  • p j represents the transposition of the coordinates of the known point P in the detection device coordinate system
  • p j refers to the point cloud data obtained by the detection scanner
  • p j Substitute into formula (1) to obtain the coordinates of the known point P in the machine tool coordinate system. This is because the machine tool needs to issue a control command based on its own coordinate system, so it is necessary to convert the data obtained based on the detection device coordinate system to the machine tool available parameters. ).
  • Equation (1) after the fact, is a rotating coordinate transformation to a translation, in FIG. 2 X e, Y e, Z e represents the axis that is the first point cloud data coordinate system after the rotation.
  • the scanning actuator is used to scan and detect the chuck 3, and the acquired point cloud data of the chuck is used to calibrate the angle information of the chuck rotation;
  • a convex calibration part 30 is installed on the edge of the chuck 3 so that the point cloud data generated by scanning can determine the rotation information of the chuck. To match the calibration.
  • Step S3 Perform data processing on the workpiece point cloud data in the machine tool coordinate system, including:
  • Step S31 the subject point cloud clustering extraction method is used to remove large noise points outside the workpiece point cloud data in the machine tool coordinate system to obtain the original point cloud data of the blade model;
  • Step S32 using the statistical outlier method to remove the original data noise points in the original point cloud data of the blade model to obtain noise-reduced point cloud data;
  • Step S33 the bounding box method considering the curvature is used to simplify and smooth the noise-reduced point cloud data
  • the original point cloud data P i needs to be roughly processed, and the main point cloud cluster extraction method is used to obtain the original point cloud data of the blade model to remove large noise points in vitro.
  • the subject point cloud clustering extraction method is introduced.
  • the point clouds are clustered according to the density, and the point cloud set with the highest density is the blade.
  • the point cloud data of the subject is denoted as MP, and the set with a lower density is denoted as OP i .
  • the distance between OP i and MP is calculated according to formula (2). If the distance exceeds the set threshold D ⁇ , it indicates that OP i is an in vitro that can be deleted.
  • k represents the number of point clouds in the neighborhood M of the point P i and M j represents the j-th point in the neighborhood M.
  • the purpose of streamlining and smoothing is to remove redundant data and make point cloud data smoother while maintaining the curvature characteristics of the workpiece.
  • the present invention introduces a bounding box method that considers the curvature characteristics.
  • the idea of the bounding box method is to put the noise-reduced point cloud data into a cuboid, and divide the cuboid into equilateral cubes according to the reduction rate. For each cube, select the point cloud data closest to the center of the cube as the feature point.
  • this method is prone to loss of features in point cloud data with large curvature (K>0.33), especially at high curvatures of the blade edge of aviation blades.
  • the present invention considers the curvature characteristics, and reduces the side length of the cube and increases the density of the area with a large curvature.
  • calculate the curvature using the quadratic surface fitting method that is, first fit the quadratic surface f(x, y) to the K adjacent points of any point P i in the point cloud according to formula (4), and calculate f(x, y)
  • the principal curvature and principal direction of determine the curvature of P i according to the knowledge of differential geometry, solve the first-order partial derivative and second-order partial derivative of the quadric surface, respectively: f x , f y , f xx , f xy , f yy ; then point Mean curvature of P i It can be obtained by formula (5);
  • the point cloud data after the above step 3 can be shown in Figure 4.
  • the processed point cloud data can still effectively maintain the key characteristics of the blade, especially the most important leaf edge characteristics for surface strengthening processing .
  • Step S4 Obtain the workpiece shape error characteristics based on the theoretical design data of the workpiece to be processed and the workpiece point cloud data in the machine tool coordinate system after data processing, and compensate the workpiece shape error characteristics;
  • the strengthening process does not change the blade size or profile, nor can it correct the error. Therefore, the error in the blade strengthening process mainly refers to the theoretical position of the blade and the actual position on the machine tool.
  • the error between the former is obtained according to the blade design model, and the latter is obtained after processing the online point cloud data, and then the above error can be obtained by analysis.
  • the typical error is shown in Figure 5.
  • the figure is a schematic diagram of a section of the blade in the XY plane.
  • the circle represents the position of the chuck 3, which can reflect the center position of the machine tool.
  • the arrow represents the processing head 6 on both sides of the blade.
  • the line indicates the cross-sectional position according to the blade design model, and the dotted line indicates the actual blade cross-sectional position.
  • Fig. 5 due to the manufacturing error of the machine tool and the blade on a certain processing section, there may be errors in the X direction, Y direction and rotation direction at the same time, and the shape and size of the section will also be different.
  • both sides need to be processed at the same time and processed to the specified section area.
  • the present invention analyzes the workpiece point cloud data (that is, the actual position data of each section of the workpiece 2 on the X axis and the Y axis) and the theoretical design data of the processed workpiece (that is, the data processing of the machine tool coordinate system) ,
  • the maximum deviation value in the X direction and Y direction between the design position data on the X axis and the Y axis of the cross section of each layer of the workpiece 2 to be processed, as an error feature, and force feedback is used for the error in the X direction
  • the compensation method performs error compensation, and increases the deviation of the trajectory for the error in the Y direction (in the example of the present invention, the actual trajectory refers to the actual trajectory that needs to be processed (ie, the data composed of the scanned and processed point cloud), and the theory The trajectory refers to the data given by the three-dimensional model.
  • the trajectory deviation value here refers to an adjustment value given to the machine tool. Increasing this value can make the machine tool move from the theoretical trajectory to the actual trajectory. Ultrasonic rolling strengthens the processing trajectory and increases or decreases the trajectory offset of the processing head 6 in the Y direction) for error compensation. Specifically, due to the error in the X direction, the preload of the processing head 6 on both sides and the surface of the workpiece are different (the preload will be generated when the processing head 6 is pressed against the surface of the workpiece).
  • the force sensor installed on the processing head 6 can Adjust the amount of expansion and contraction of the internal spring of the processing head 6 according to the real-time change of the pre-pressure, so as to adjust the pre-pressure of the processing head 6 and the workpiece surface to make it constant, and then achieve the effect of compensating the error in the X direction; for example, as shown in FIG. 6 and FIG. 6
  • the solid line represents the actual position of the blade surface
  • the dashed line represents the theoretical position of the blade surface.
  • the processing head 6 that follows the dashed line according to the theoretical value needs to be adjusted to the actual blade solid line position. A in FIG.
  • the processing head 6 shows the processing head 6 Processing according to the theoretical dashed line, so the pre-pressure is smaller than the processing surface, b indicates the position of the processing head 6 after compensation, which has been fitted to the actual blade surface, and c indicates that the processing continues after compensation, and the processing head 6 can always stick If the blade is not closed at this position, the pre-pressure will be too large.
  • the error of the workpiece in the Z-axis rotation direction is less affected.
  • the method of increasing the trajectory deviation value can also be used to compensate the error in the Z direction.
  • the present invention stops strengthening processing during detection scanning, and analyzes the processed model of the point cloud data obtained by scanning and the design model of the workpiece to adjust the trajectory of strengthening processing, thereby achieving the purpose of error compensation.
  • the on-machine detection design scheme proposed by the present invention can be designed according to the structural characteristics of the ultrasonic rolling machine, which simplifies the design difficulty of the scanning actuator; after comprehensive analysis and comparison of several different scanners, the selected
  • the laser scanner has the advantages of good point cloud acquisition effect, low cost, and appropriate size;
  • the proposed device calibration method is simple and practical, and has strong feasibility;
  • the proposed large-scale complex curved point cloud processing method is accurately obtained
  • the key to surface error and later compensation is based on the characteristics of surface enhancement of complex surfaces.
  • the error is analyzed from the final results to simplify the difficulty of feature analysis.
  • the proposed error compensation scheme is in line with the actual situation of surface enhancement processing of complex surfaces.
  • the on-machine detection method for the surface strengthening processing of complex curved workpieces uses advanced non-contact rapid shape detection technology. Compared with conventional detection methods, it has the characteristics of no disassembly and assembly on the machine, fast speed, high precision, and strong technology. Advanced and innovative.
  • the large-scale point cloud processing technology of complex curved surfaces studied in the present invention is more challenging than the point cloud processing problem of general workpieces; it is mainly reflected in the characteristics of irregular surface shapes and large changes in shape curvature, and requires accuracy High and computationally intensive. Therefore, it has higher research difficulty.
  • the proposed subject point cloud clustering extraction, statistical outlier method for denoising, and curvature-enclosed box method for streamlining and smoothing have distinct characteristics.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Mechanical Engineering (AREA)
  • Optics & Photonics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Length Measuring Devices With Unspecified Measuring Means (AREA)
  • Numerical Control (AREA)

Abstract

一种复杂表面加工的在机点云检测及补偿方法，其包括：步骤S1，在超声滚压机床上安装一检测扫描执行机构；步骤S2，利用所述扫描执行机构对所述被加工工件（2）进行扫描，以获得检测设备坐标系下的工件点云数据，并将该检测设备坐标系下的工件点云数据转换为机床坐标系下的工件点云数据；步骤S3，对所述机床坐标系下的工件点云数据进行数据处理；步骤S4，根据所述被加工工件（2）的理论设计数据以及经过数据处理的所述机床坐标系下的工件点云数据，获得工件形状误差特征，并对该工件形状误差特征进行补偿。

Description

一种复杂表面加工的在机点云检测及补偿方法 技术领域
本发明涉及一种复杂表面加工的在机点云检测及补偿方法。
背景技术
在复杂曲面工件表面，例如飞机发动机叶片表面的强化过程中，研究发现，经超声滚压机床强化加工的叶片区域与所规划的加工区域之间存在一定的偏差，这是由于叶片加工成型的形状尺寸误差、机床综合误差及卡盘安装误差，导致了叶片的实际形状位置与规划时采用的模型形状位置存在差异。因此，在表面强化的过程中，基于复杂曲面设计模型所规划的加工轨迹往往会造成加工误差，即所强化复杂曲面区域与目标区域之间存在误差。这就需要人工调节参数，修正加工轨迹，从而补偿加工误差。然而，由于每个复杂曲面工件的误差各不相同，这通常需要在实际加工中暂停设备，进行补偿参数的调节，大大降低了加工效率，增加了制造成本。而且人工调节要求经验高，难度大，精度控制难度高。
为此，现有技术中通常采用常规检测设备(例如，三坐标测量机、自动绘图测量仪等)对工件复杂曲面型面进行测量，即，通过检测设备获取复杂曲面型面上离散点的坐标数据实现其型面的评定。基于上述常规检测设备，现有的复杂曲面工件的测量方法主要包括：标准样板法、自动绘图测量法、光学投影测量法、坐标测量法和电感测量法等，其中：标准样板法是一种定性的测量方法，只能凭借经验判断零件是否合格，无法给出工件实际误差的大小；自动绘图测量法的测量精度虽然较标准样板法高，但是仍然无法满足高精度叶片检测的需要，且自动绘图测量仪体积庞大、操作不便，因此应用范围有限；光学投影测量法则一般用于定性检测，通过观察实际叶型与理论叶型的符合程度直接判断叶片是否合格，但是在通过读取坐标值进行定量测量时，效率比较低下，且其受自身显示屏限制，测量范围有限，同时其受工件表面反射能力影响较大，测量精度较低；坐标测量法作为一种通用的点测量设备，造价高、测量过程繁琐、测量效率较低、需要二次开发测量软件、对环境要求高；电感测量法则由于属于非完全轮廓检测，使得测量结果不是十分可靠，有误判的可能性，并且对不同工件需要配备相应的支架装置和标 准工件，增加了测量成本和时间。
近几年，光学测量技术的发展为复杂曲面型面测量提供了新的测量方法，人们可以利用先进的复杂曲面形状检测设备，例如金属表面缺陷检测仪；三维扫描仪(包括激光三维扫描仪、结构光三维扫描仪，实现对工件表面的精确检测与误差评估。基于上述先进的检测设备，新的测量方法主要包括：机器视觉测量法、四坐标激光测量法和激光扫描测量法等，这些测量方法的优点在于测量速度快、范围大、效率高。
然而，上述的这些现有的和改进的检测方法通常都要求将工件安装在检测台上，这就造成了二次装卸带来的误差。
为了能够在加工过程中进行检测和补偿误差，必须将检测与加工进行集成，成为“加工-集成”系统。因此，针对加工中心的在机检测也得到了广泛的应用。国外很多公司对这个方面进行了研究，并针对各种不同的制造设备，研制出符合各自需要的在机检测系统。例如，英国Reni Shaw公司专门开发的安装在加工中心的各类测头和对应的软件，可以实现对刀、刀具破损检测、工件误差检测等功能；德国Kapp公司磨齿机的在机齿轮测量装置；著名的齿轮机床制造商格里森公司及克林格贝尔公司则将齿轮检测中心与加工机床相连，实现了齿轮的CAD/CAM/CAI的闭环制造；英国Delcam公司的通用检测系统Power INSPECT可以实现在机检测和脱机检测。国内学者也相继开展了在机检测技术方面的探索和研究工作。相关的研究主要包括：机床的误差综合建模，误差参数测量及辨识，工件检测路径自动规划，以及针对不同误差源的补偿技术。然而，现有的误差补偿方法，如基于反变形误差多次补偿方法，具有误差补偿效率低下等缺点。
在机检测设备的研发可选择接触式的检测探头或者基于光学原理的非接触式检测探头，其中，接触式的检测探头直接与待测物体接触，触发测头传感器发送测量信号，接收传感器接收信号，确定测头当前球心坐标，经过测头半径补偿和误差补偿最终计算出该点的坐标值，其具有精度高、环境适应性强的特点，但存在检测效率低、检测成本高的缺点；非接触式的检测探头则利用光学等物理特性来反求物体的形貌信息，相比接触式的检测方法，具有扫描速度快的特点，能获取被测工件的整体轮廓，且能测量表面较软的工件，精度较高，可达0.01mm精度，但同时也容易受到工件表面反射特性(如曲率、颜色、粗糙程度等)的影响。
针对航空薄壁叶片这类自由曲面，通常需多次反复测量，导致测量点云规模达十万/百万级；更为严重的是，受工业现场温度差、灰尘颗粒、微小振动等因素影响，测量点云存在固有缺陷(残余噪音、体外孔洞、层叠区域、密度差异等)。由此带来的大规模测量点云数据处理则成为复杂曲面检测的关键问题之一，如精简、光顺、拼合和重构等几何操作；具体来说，由于航空叶片前后缘区域壁厚不足2毫米，常规精简或光顺过程容易丢失前后缘数据；由于叶片点云本身为非均匀分布，在精简过程中要在点云分布稀疏区域适当保留稀疏点云，否则会造成空洞；另外还要避免过光顺导致的丢失叶缘高曲率特征。
因此，为了提高复杂曲面工件表面强化加工效率与精度，需要研发一种可以在复杂曲面工件在机加工工位进行的快速高效的检测技术，以及高效补偿表面强化误差的方法。
发明内容
为了解决上述现有技术存在的问题，本发明旨在提供一种复杂表面加工的在机点云检测及补偿方法，以在超声滚压强化加工过程中，高效快速地在机检测及补偿误差。
本发明所述的一种复杂表面加工的在机点云检测及补偿方法，其包括以下步骤：
步骤S1，在超声滚压机床上安装一检测扫描执行机构，其中，所述超声滚压机床包括：一卡盘以及两个分别安装在所述卡盘左右两侧的加工头；所述检测扫描执行机构包括：两个分别安装在所述加工头上的检测扫描器；
步骤S2，利用所述扫描执行机构对安装在所述卡盘上的被加工工件进行扫描，以获得检测设备坐标系下的工件点云数据，并将该检测设备坐标系下的工件点云数据转换为机床坐标系下的工件点云数据；
步骤S3，对所述机床坐标系下的工件点云数据进行数据处理；
步骤S4，根据所述被加工工件的理论设计数据以及经过数据处理的所述机床坐标系下的工件点云数据，获得工件形状误差特征，并对该工件形状误差特征进行补偿；
其中，所述步骤S3包括：
步骤S31，采用主体点云聚类提取法去除机床坐标系下的工件点云数据中 的体外噪声点，以获取叶片模型原始点云数据；
步骤S32，利用统计离群点法去除叶片模型原始点云数据中的原始数据噪声点，以获取降噪后的点云数据；
步骤S33，采用考虑曲率的包围盒法对降噪后的点云数据进行精简和光顺。
在上述的复杂表面加工的在机点云检测及补偿方法中，所述步骤S31包括：
将机床坐标系下的工件点云数据按照密度大小进行聚类，其中密度最大的点云集合为叶片的主体点云数据，记为MP，密度小的点云集合记为OP i，根据以下公式计算OP i到MP的距离，若该距离超过一设定阈值D τ，则表示OP i为待删除的所述体外噪声点；
Dist(OP i，MP)＝||OP i-MP||。
在上述的复杂表面加工的在机点云检测及补偿方法中，所述步骤S32包括：
根据以下公式计算所述叶片模型原始点云数据中各个点P i到邻域M内所有点的平均距离d mid(P i)，若d min(P i)＞D mid，则点P i被定义为离群点，最后将离群点删除，其中，D mid为模型的全局平均距离；
Figure PCTCN2019080138-appb-000001
其中，k表示点P i的邻域M中的点云数目，M j表示邻域M中第j个点。
在上述的复杂表面加工的在机点云检测及补偿方法中，所述步骤S33包括：把所述降噪后的点云数据放入到一个长方体中，根据预设的精简率的大小将该长方体均匀分割成等边的立方体，对于每个立方体选取距离该立方体中心最近的点云数据作为特征点，并且对于所述降噪后的点云数据中曲率大于0.33的区域，缩小所述立方体的边长，以增大其密度。
在上述的复杂表面加工的在机点云检测及补偿方法中，所述步骤S33还包括：利用二次曲面拟合法计算所述降噪后的点云数据的曲率。
在上述的复杂表面加工的在机点云检测及补偿方法中，所述工件形状误差特征包括：所述被加工工件的每层截面在X轴和Y轴上的实际位置数据与所述被加工工件的每层截面在X轴和Y轴上的设计位置数据之间在X方向和Y方向上的最大偏差值，所述步骤S4包括：采用力反馈补偿法，对所述在X 方向上的最大偏差值进行补偿，以及通过增加或减少所述加工头在Y方向上的轨迹偏移量，对所述在Y方向上的最大偏差值进行补偿。
在上述的复杂表面加工的在机点云检测及补偿方法中，所述超声滚压机床还包括：一床身底座，所述卡盘可绕Z轴旋转并可沿Z轴移动地安装在所述床身底座上。
在上述的复杂表面加工的在机点云检测及补偿方法中，所述检测扫描执行机构还包括：两个安装在所述床身底座上并对称分布在所述被加工工件左右两侧的Y轴升降移动工作台以及两个分别安装在所述Y轴升降移动工作台上的X轴左右移动工作台。
在上述的复杂表面加工的在机点云检测及补偿方法中，所述检测扫描器为激光扫描器。
由于采用了上述的技术解决方案，本发明为了避免复杂曲面工件在检测中二次装卸带来的检测误差，增加检测的效率，利用非接触式检测技术在复杂曲面工件表面强化过程中进行工件的在机检测，即，在复杂曲面工件加工工位对工件曲面表面进行扫描，快速获取复杂曲面形状数据，并通过大规模点云处理技术，获得该复杂曲面形状位置的综合误差，通过分析点云数据模型与设计模型之间差异的特征，得到基于力反馈与位置补偿器的误差补偿方案(即，向机床子带的位置补偿器输出补偿参数)，进行轨迹误差补偿，从而提高了复杂曲面表面强化加工的精度和效率。
附图说明
图1a是本发明一种复杂表面加工的在机点云检测及补偿方法中检测扫描执行机构安装在超声滚压机床上的结构示意图；
图1b是本发明中加工工件、卡盘、加工头和检测扫描器的结构简图，其中示出了卡盘和检测扫描器的X、Y、Z轴运动方向；
图2是本发明中检测设备坐标系与机床坐标系的示意图；
图3是本发明中原始点云数据的示意图；
图4是本发明中经过步骤S3处理后的点云数据的示意图
图5是本发明中叶片某一加工截面的X-Y平面上的误差特征分析示意图；
图6是本发明中采用力反馈补偿法补偿X方向上的误差的示意图。
具体实施方式
下面结合附图，给出本发明的较佳实施例，并予以详细描述。
本发明，即一种复杂表面加工的在机点云检测及补偿方法，包括以下步骤：
步骤S1，在超声滚压机床上安装检测扫描执行机构，其中，
如图1a、1b所示，超声滚压机床包括：安装在床身底座1上并供被加工工件2安装于其上的卡盘3，其中，卡盘3可绕Z轴旋转并可在床身底座1上沿Z轴移动，被加工工件2的末端安装在卡盘3的一侧端面上，即，通过卡盘3可带动被加工工件1绕Z轴旋转以及沿Z轴方向进给；
扫描执行机构包括：两个安装在床身底座1上并对称分布在被加工工件2左右两侧的Y轴升降移动工作台4、两个分别安装在Y轴升降移动工作台4上的X轴左右移动工作台5、两个分别安装X轴左右移动工作台5上的加工头6(加工头6为超声滚压机床的固有部件)以及两个分别安装在加工头6上的检测扫描器7，即，通过Y轴升降移动工作台4和X轴左右移动工作台5可带动检测扫描器7沿Y轴方向上下移动以及沿X轴方向左右移动；
在本实施例中，被加工工件1为飞机发动机叶片(发动机叶片种类繁多，其叶型多呈空间自由曲面状；本发明适用于双圆弧、多圆弧或S叶型等类型叶片，且最大叶身长度为1200mm)；由于叶片表面反光的影响问题，通过分析比较激光扫描器与结构光扫描器对复杂曲面工件扫描生成的点云效果，综合扫描器成本、尺寸以及扫描器要求，最终选取激光扫描器作为检测扫描器7以用于叶片点云获取，具体来说，结构光扫描器是通过拍照的方式提取物体表面三维数据，而激光扫描器是通过线扫描提取物体表面三维数据，对于复杂曲面工件，尤其是航空发动机叶片，其金属表面光泽且反光，会严重影响结构光扫描器获取点云的效果，而用激光扫描器可以避免这一问题，而且激光扫描器获得的点云比结构光扫描器获取的点云精度高，即，获取点云数据的质量高，另外，激光扫描器比结构光扫描器更加便携，可以在操作现场进行任何角度的无死角扫描；在本实施例中，激光扫描器的尺寸约在20mm*30mm*10，并且其工作特点是：固定式全自动扫描，效率高，精度高，且分辨率高。
步骤S2，利用扫描执行机构，即利用检测扫描器沿Y轴方向的上下移动 以及沿X轴方向的左右移动，并结合卡盘3在Z轴方向上的进给和绕Z轴的旋转，对被加工工件2进行扫描，以获得检测设备坐标系下的工件点云数据，并将该检测设备坐标系下的工件点云数据转换为机床坐标系下的工件点云数据；
具体来说，由于检测设备坐标系与机床坐标系不同，需要进行坐标系转换，并进行参数标定(参数标定是对安装检测机构及每一次装夹待加工件后，相对机床绝对坐标系进行定位，标定的旋转角度信息将保证检测开始时处于真实零点位置)。
如图2所示，根据公式(1)可将检测设备坐标系转换为机床坐标系：
Figure PCTCN2019080138-appb-000002
其中，P i表示机床坐标系，P j表示检测设备坐标系，
Figure PCTCN2019080138-appb-000003
表示检测设备坐标系与机床坐标系的原点三轴的相对平移位置，
Figure PCTCN2019080138-appb-000004
为旋转矩阵，表示检测设备坐标系相对初始位置旋转量，p j表示已知点P在检测设备坐标系下的坐标的转置(p j是指通过检测扫描器获得的点云数据，将p j代入公式(1)即可获得机床坐标系下已知点P的坐标，这是因为机床需要基于自己的坐标系发出控制指令，故需将基于检测设备坐标系得到的数据转换为机床可用参数)。
由公式(1)可以看出，坐标系转换其实是先旋转后平移，图2中的X e、Y e、Z e轴即表示将点云数据先进行旋转后的坐标系。
由于卡盘3的形状为圆形，位置固定，因此，先利用扫描执行机构对卡盘3进行扫描检测，并利用获取到的卡盘的点云数据进行卡盘旋转的角度信息的标定；在本实施例中，由于卡盘3为圆形，无法识别旋转信息，因此在卡盘3的边缘安装一个凸起的标定零件30，使得扫描生成的点云数据中可以判断卡盘的旋转信息，以配合标定。
步骤S3，对机床坐标系下的工件点云数据进行数据处理，包括：
步骤S31，采用主体点云聚类提取法去除机床坐标系下的工件点云数据中的体外大片噪声点，以获取叶片模型原始点云数据；
步骤S32，利用统计离群点法去除叶片模型原始点云数据中的原始数据噪声点，以获取降噪后的点云数据；
步骤S33，采用考虑曲率的包围盒法对降噪后的点云数据进行精简和光顺；
具体来说，将通过步骤S2获得的原始的机床坐标系下的工件点云数据记为原始点云数据P i＝{x i，y i，z i}。由于受测量设备的精度、扫描速度、被测量零件的表面质量以及外界环境等因素的影响，不可避免地在点云数据中引入噪声点。这些噪声点的存在，将直接影响被测工件的精度。因此必须对上述原始点云数据P i进行去噪处理。
首先，需要对原始点云数据P i进行粗处理，采用主体点云聚类提取的方法获取叶片模型原始点云数据，去除体外大片噪声点。
如图3所示，当叶片点云模型体外非连接噪声点较多时，引入主体点云聚类提取法，首先将点云按照密度大小进行聚类，其中密度最大的点云集合便是叶片的主体点云数据，记为MP，密度较小的集合记为OP i，根据公式(2)计算OP i到MP的距离，若距离超过设定阈值D τ，则表示OP i是可以删除的体外非连接噪声点；
Dist(OP i，MP)＝||OP i-MP||    (2)。
然后，进行精处理，利用统计离群点法彻底去除原始数据噪声点，即，运用统计学的方法对每个点的邻域进行分析，包括：首先根据公式(3)计算该点P i到邻域M内所有点的平均距离d mid(P i)，模型的全局平均距离D mid符合高斯分布D～N(μ，σ 2)，其特征由均值μ和标准差σ决定。若d min(P i)＞D mid，则P i点被定义为离群点，最后将离群点删除；
Figure PCTCN2019080138-appb-000005
其中，k表示点P i的邻域M中的点云数目，M j表示邻域M中第j个点。
最后，进行精简和光顺的目的在于：在保持工件曲率特征的前提下，去除冗余数据并使点云数据更加平滑。为此，本发明引入考虑曲率特征的包围盒法，该包围盒法的思想是把降噪后的点云数据放入到一个长方体中，根据精简率的大小将长方体均匀分割成等边小立方体，对于每个立方体选取距离立方体中心最近的点云数据作为特征点。但是这种方法在曲率较大(K＞0.33)的点云数据中，易导致特征丢失，尤其对于航空叶片叶缘高曲率处。因此本发明考虑曲率特征，对于曲率较大的区域缩小立方体边长，增大其密度。利用二次曲面拟合法计算曲率，即，首先根据公式(4)对点云中任意一点P i的K个临近点拟合二次曲面f(x，y)，通过计算f(x，y)的主曲率及主方向确定 P i的曲率；根据微分几何知识求解二次曲面一阶偏导和二阶偏导，分别是：f x，f y，f xx，f xy，f yy；则点P i的平均曲率
Figure PCTCN2019080138-appb-000006
可由公式(5)求得；
f(x，y)＝ax 2+bxy+cy 2      (4)，
Figure PCTCN2019080138-appb-000007
其中
Figure PCTCN2019080138-appb-000008
是点P i的单位法向量，计算如下：
Figure PCTCN2019080138-appb-000009
经过上述步骤3后的点云数据可如图4所示，从图4可以看出，处理后的点云数据仍能有效保持叶片的关键特征，尤其是对于表面强化加工最重要的叶缘特征。
步骤S4，根据被加工工件理论设计数据以及经过数据处理的机床坐标系下的工件点云数据，获得工件形状误差特征，并对该工件形状误差特征进行补偿；
具体来说，以飞机叶片为例，其强化加工并非改变叶片尺寸或型面，也不能纠正误差，因此，叶片强化加工中的误差主要是指叶片理论上的位置与在机床上的实际位置之间的误差，前者根据叶片设计模型获得，后者根据在线检测点云数据经处理后获得，然后，可以分析获得以上所述误差。典型的误差如图5所示，图中为X-Y平面中叶片某一截面的示意图，其中圆形代表卡盘3的位置，其可反映机床中心位置，箭头代表叶片两侧的加工头6，实线表示根据叶片设计模型所在的截面位置，而虚线则表示实际的叶片截面位置。由图5可见，在某一个加工截面上，由于机床及叶片的制造误差，截面可能同时存在X方向、Y方向以及旋转方向的误差，并且截面形状尺寸也会有所不同。对于叶片的超声滚压强化而言，在每一个加工截面上，需要两侧同时加工，并加工到指定的截面区域。图5上所示的各种误差最终将导致以下结果：一方面是造成左右两侧加工头6与叶片截面接触位置错误，另一方面是造成两侧加工头6在Y方向上上端超出叶片截面区域，下端无法达到最下端位置。
因此，本发明分析经过数据处理的机床坐标系下的工件点云数据(即， 被加工工件2的每层截面在X轴和Y轴上的实际位置数据)与被加工工件理论设计数据(即，被加工工件2的每层截面在X轴和Y轴上的设计位置数据)之间在X方向和Y方向上的最大偏差值，作为误差特征，并且对于在X方向上的误差采用力反馈补偿法进行误差补偿，对于在Y方向上的误差采用增加轨迹偏差值(在本发明实例中，实际轨迹是指实际需要加工的轨迹(即扫描及处理出来的点云构成的数据)，而理论轨迹是指三维模型给出的数据，此处的轨迹偏差值是指赋予机床的一个调整值，增加这个值即可使机床从理论轨迹走向实际轨迹)的方法(即，在强化加工时通过调整超声滚压强化加工轨迹，增加或减少加工头6在Y方向上的轨迹偏移量)进行误差补偿。具体来说，由于X方向上的误差造成两侧加工头6与工件表面的预压力不同(当加工头6顶在工件表面时会产生预压力)，利用安装在加工头6上的力传感器可以根据预压力的实时变化调整加工头6内部弹簧的伸缩量，从而调整加工头6与工件表面的预压力使之恒定，进而达到补偿X方向上误差的效果；例如图6所示，图6中实线表示叶片表面的实际位置，虚线表示叶片表面的理论位置，加工时，需要将按理论值走虚线的加工头6，调整到实际叶片的实线位置，图6中a示意为加工头6按理论虚线加工，所以脱离加工表面，为预压力偏小，b示意为经补偿后加工头6的位置，已贴合实际叶片表面，c示意为经补偿后继续加工，加工头6能始终贴合叶片，此位置若不补偿，则会预压力过大。另外，根据实际的加工要求和经验，工件在Z轴旋转方向上的误差影响较小，当然，也可采用增加轨迹偏差值的方法进行Z方向上的误差补偿。需要注意的是，本发明在检测扫描时停止强化加工，通过分析扫描获得的点云数据经处理后的模型和工件设计模型，来调整强化加工的轨迹，进而达到误差补偿的目的。
综上所述，本发明所提出的在机检测的设计方案能够根据超声滚压机床的结构特点进行设计，简化了扫描执行机构的设计难度；经综合分析比较几种不同的扫描器，所选用的激光扫描器具有点云获取效果好，成本低，外形尺寸适当等优点；所提出的设备标定方法简单实用，具有很强的可行性；所提出的大规模复杂曲面点云的处理方法是准确获得曲面误差和后期补偿的关键，根据复杂曲面表面强化的特点，从最终结果上对误差进行了特征分析，简化了特征分析难度；所提出的误差补偿方案符合复杂曲面表面强化加工中 的实际情况。该复杂曲面工件表面强化加工的在机检测方法运用先进的非接触式快速形状检测技术，相比常规的检测方法，具有在机免拆装，速度快，精度高的特点，具有很强的技术先进性和创新性。同时，本发明中所研究的复杂曲面大规模点云处理技术，相比于一般工件的点云处理问题更具挑战性；主要体现在表面形状不规则，形状曲率变化大的特点，且要求精度高，计算量大。因此，具有较高的研究难度。另外，所提出的主体点云聚类提取、统计离群点法去噪、考虑曲率的包围盒法进行精简和光顺，具有鲜明的特色。
以上所述的，仅为本发明的较佳实施例，并非用以限定本发明的范围，本发明的上述实施例还可以做出各种变化。凡是依据本发明申请的权利要求书及说明书内容所作的简单、等效变化与修饰，皆落入本发明专利的权利要求保护范围。本发明未详尽描述的均为常规技术内容。

Claims (9)

  1. 一种复杂表面加工的在机点云检测及补偿方法，其特征在于，所述方法包括以下步骤：
    步骤S1，在超声滚压机床上安装一检测扫描执行机构，其中，所述超声滚压机床包括：一卡盘以及两个分别安装在所述卡盘左右两侧的加工头；所述检测扫描执行机构包括：两个分别安装在所述加工头上的检测扫描器；
    步骤S2，利用所述扫描执行机构对安装在所述卡盘上的被加工工件进行扫描，以获得检测设备坐标系下的工件点云数据，并将该检测设备坐标系下的工件点云数据转换为机床坐标系下的工件点云数据；
    步骤S3，对所述机床坐标系下的工件点云数据进行数据处理；
    步骤S4，根据所述被加工工件的理论设计数据以及经过数据处理的所述机床坐标系下的工件点云数据，获得工件形状误差特征，并对该工件形状误差特征进行补偿；
    其中，所述步骤S3包括：
    步骤S31，采用主体点云聚类提取法去除机床坐标系下的工件点云数据中的体外噪声点，以获取叶片模型原始点云数据；
    步骤S32，利用统计离群点法去除叶片模型原始点云数据中的原始数据噪声点，以获取降噪后的点云数据；
    步骤S33，采用考虑曲率的包围盒法对降噪后的点云数据进行精简和光顺。
  2. 根据权利要求1所述的复杂表面加工的在机点云检测及补偿方法，其特征在于，所述步骤S31包括：
    将机床坐标系下的工件点云数据按照密度大小进行聚类，其中密度最大的点云集合为叶片的主体点云数据，记为MP，密度小的点云集合记为OP i，根据以下公式计算OP i到MP的距离，若该距离超过一设定阈值D τ，则表示OP i为待删除的所述体外噪声点；
    Dist(OP i，MP)＝||OP i-MP||。
  3. 根据权利要求1所述的复杂表面加工的在机点云检测及补偿方法，其特征在于，所述步骤S32包括：
    根据以下公式计算所述叶片模型原始点云数据中各个点P i到邻域M内所有点的平均距离d mid(P i)，若d min(P i)＞D mid，则点P i被定义为离群点，最后 将离群点删除，其中，D mid为模型的全局平均距离；
    Figure PCTCN2019080138-appb-100001
    其中，k表示点P i的邻域M中的点云数目，M j表示邻域M中第j个点。
  4. 根据权利要求1所述的复杂表面加工的在机点云检测及补偿方法，其特征在于，所述步骤S33包括：把所述降噪后的点云数据放入到一个长方体中，根据预设的精简率的大小将该长方体均匀分割成等边的立方体，对于每个立方体选取距离该立方体中心最近的点云数据作为特征点，并且对于所述降噪后的点云数据中曲率大于0.33的区域，缩小所述立方体的边长，以增大其密度。
  5. 根据权利要求4所述的复杂表面加工的在机点云检测及补偿方法，其特征在于，所述步骤S33还包括：利用二次曲面拟合法计算所述降噪后的点云数据的曲率。
  6. 根据权利要求1所述的复杂表面加工的在机点云检测及补偿方法，其特征在于，所述工件形状误差特征包括：所述被加工工件的每层截面在X轴和Y轴上的实际位置数据与所述被加工工件的每层截面在X轴和Y轴上的设计位置数据之间在X方向和Y方向上的最大偏差值，所述步骤S4包括：采用力反馈补偿法，对所述在X方向上的最大偏差值进行补偿，以及通过增加或减少所述加工头在Y方向上的轨迹偏移量，对所述在Y方向上的最大偏差值进行补偿。
  7. 根据权利要求1所述的复杂表面加工的在机点云检测及补偿方法，其特征在于，所述超声滚压机床还包括：一床身底座，所述卡盘可绕Z轴旋转并可沿Z轴移动地安装在所述床身底座上。
  8. 根据权利要求7所述的复杂表面加工的在机点云检测及补偿方法，其特征在于，所述检测扫描执行机构还包括：两个安装在所述床身底座上并对称分布在所述被加工工件左右两侧的Y轴升降移动工作台以及两个分别安装在所述Y轴升降移动工作台上的X轴左右移动工作台。
  9. 根据权利要求1所述的复杂表面加工的在机点云检测及补偿方法，其特征在于，所述检测扫描器为激光扫描器。
PCT/CN2019/080138 2018-12-10 2019-03-28 一种复杂表面加工的在机点云检测及补偿方法 Ceased WO2020118970A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/312,543 US12061078B2 (en) 2018-12-10 2019-03-28 On-machine inspection and compensation method employing point clouds and applied to complex surface processing

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811505485.4 2018-12-10
CN201811505485.4A CN109489580B (zh) 2018-12-10 2018-12-10 一种航空发动机叶片表面加工的在机点云检测及补偿方法

Publications (1)

Publication Number Publication Date
WO2020118970A1 true WO2020118970A1 (zh) 2020-06-18

Family

ID=65709746

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/080138 Ceased WO2020118970A1 (zh) 2018-12-10 2019-03-28 一种复杂表面加工的在机点云检测及补偿方法

Country Status (3)

Country Link
US (1) US12061078B2 (zh)
CN (1) CN109489580B (zh)
WO (1) WO2020118970A1 (zh)

Cited By (25)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111798570A (zh) * 2020-07-16 2020-10-20 浙江大学 一种叶轮机械叶片曲面网格化快速生成方法
CN112215841A (zh) * 2020-10-31 2021-01-12 上海扩博智能技术有限公司 风机叶片点云数据的形成方法、系统、设备和存储介质
CN112581579A (zh) * 2020-12-22 2021-03-30 同济大学 一种磁浮滑行面点云数据提取方法
CN112626321A (zh) * 2020-12-28 2021-04-09 华南理工大学 Rv减速器摆线轮超声滚压过程中的变形控制装置及方法
CN112731864A (zh) * 2020-12-21 2021-04-30 北京理工大学 一种机床加工误差补偿方法、装置及零件加工机床
CN112762856A (zh) * 2020-12-29 2021-05-07 中国航空工业集团公司西安飞机设计研究所 一种用于扫描影响因素的评价方法
CN113369998A (zh) * 2021-06-30 2021-09-10 中国航发动力股份有限公司 一种基于工艺模型的模锻叶片型面自适应补偿加工方法
CN113390340A (zh) * 2021-06-15 2021-09-14 中国工程物理研究院机械制造工艺研究所 一种非连续区域的角球面球心空间位置检测方法
CN113610916A (zh) * 2021-06-17 2021-11-05 同济大学 基于点云数据的不规则物体体积测定方法及系统
CN113985813A (zh) * 2021-10-27 2022-01-28 中国航发沈阳黎明航空发动机有限责任公司 一种基于在机检测的机床原点误差补偿方法
CN114187402A (zh) * 2021-11-04 2022-03-15 河南农业大学 基于三维点云的烤烟三维模型构建方法
CN114393448A (zh) * 2022-01-21 2022-04-26 中国科学院长春光学精密机械与物理研究所 一种提高磁流变机器人抛光设备轨迹精度的方法
CN114777670A (zh) * 2022-04-21 2022-07-22 西安交通大学 一种基于接触式测头的曲面在机测量方法
CN115647438A (zh) * 2022-10-31 2023-01-31 首都航天机械有限公司 一种超大径厚比整体箱底自适应补偿加工的找正方法
CN116372668A (zh) * 2023-03-10 2023-07-04 珠海格力精密模具有限公司 一种工件加工精度检测修正方法、检测系统及电子设备
CN116512806A (zh) * 2023-04-28 2023-08-01 华侨大学 面向石雕切割工件测量的机器人扫描视点规划方法及系统
CN116952181A (zh) * 2023-09-18 2023-10-27 成都飞机工业（集团）有限责任公司 一种大长径比复合材料管型薄壁件的内型面测量方法
CN117806231A (zh) * 2024-02-27 2024-04-02 山东微晶重工有限公司 一种基于物联网的机床运行加工控制系统及方法
CN118513913A (zh) * 2024-07-19 2024-08-20 无锡康贝电子设备有限公司 一种数控机床不规则零件的尺寸测量方法
CN118623797A (zh) * 2024-05-31 2024-09-10 天津大学 一种多层结构三维信息的测量方法及装置
CN119115564A (zh) * 2024-10-16 2024-12-13 吉林大学 激光加热超声振动辅助超精密加工车铣一体机床及控制方法
CN119251292A (zh) * 2024-12-05 2025-01-03 福建省武夷山市永生茶业有限公司 用于茶陇采摘的割刀定位方法、装置、电子设备及存储介质
CN119984095A (zh) * 2025-04-11 2025-05-13 远东幕墙(珠海)有限公司 一种双曲铝板加工用测量装置
CN120235876A (zh) * 2025-05-30 2025-07-01 西安爱德华测量设备股份有限公司 一种精锻叶片进排气边双缺口叶型轮廓度检测方法
CN120791523A (zh) * 2025-08-27 2025-10-17 天津蕴宝伟业科技发展有限公司 一种无尘打磨自动生产线的板材加工控制方法及系统

Families Citing this family (42)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109489580B (zh) 2018-12-10 2021-09-24 华东理工大学 一种航空发动机叶片表面加工的在机点云检测及补偿方法
IT201900002335A1 (it) * 2019-02-18 2020-08-18 Tiberina Solutions S R L Sistema per la verifica della presenza di restrizioni di spessore su almeno un componente meccanico e relativo metodo di verifica
CN110411325B (zh) * 2019-08-14 2021-06-15 广西财经学院 一种土木工程钢筋变形测量装置
CN110940255B (zh) * 2019-12-10 2021-07-06 株洲凯丰实业股份有限公司 叶片叶型分析仪测具
CN111539074B (zh) * 2020-03-31 2023-11-14 苏州数设科技有限公司 一种工件加工数据处理方法与装置
US12380605B2 (en) * 2020-06-29 2025-08-05 Lg Electronics Inc. Point cloud data transmission device, transmission method, processing device, and processing method
CN111861941B (zh) * 2020-07-31 2024-03-22 富德康(北京)科技股份有限公司 一种三维空间测量结果数据的补偿算法
CN112548032B (zh) * 2020-12-02 2022-11-15 北京航星机器制造有限公司 一种基于三维扫描的铸件铸造方法
CN112410536B (zh) * 2020-12-07 2025-02-14 集美大学 一种超声冲击表层强化装置及方法
CN112595280B (zh) * 2020-12-21 2022-02-01 吉林大学 一种成角度复杂面形测量方法
CN113155957A (zh) * 2021-03-03 2021-07-23 广东工业大学 一种针对大变形零件内部结构曲面的超声波导向方法
CN113295092B (zh) * 2021-05-28 2022-06-14 华中科技大学 一种针对薄壁零件的线激光三维测量系统和方法
CN113781535A (zh) * 2021-08-20 2021-12-10 连云港杰瑞自动化有限公司 一种基于振动传感器的视觉3d点云数据的补偿装置及补偿方法
CN113981188A (zh) * 2021-09-23 2022-01-28 华东理工大学 一种轨迹可控双边超声滚压表面强化装置
CN114001683B (zh) * 2021-10-26 2022-09-20 大连理工大学 一种基于坐标原点平移的复杂构件加工剩余壁厚自适应补偿方法
CN114295044B (zh) * 2021-10-29 2022-11-15 重庆忽米网络科技有限公司 基于激光3d技术的汽车挡风玻璃检测方法及系统
CN114115123B (zh) * 2021-11-16 2024-04-09 上海交通大学 航空大型薄壁类非刚体零件的参数化数控加工方法和系统
CN115042015B (zh) * 2022-04-13 2023-10-20 西安交通大学 一种复杂零件关键特征参数的测头在机测量方法
CN114754699B (zh) * 2022-04-24 2023-08-25 深圳裹动科技有限公司 运动物体轮廓的构建方法及系统、以及主控设备
CN114888636B (zh) * 2022-05-09 2023-12-01 南京理工大学 一种基于三维激光扫描的刀具损伤智能化监测系统及方法
CN114820750A (zh) * 2022-06-09 2022-07-29 中科航迈数控软件(深圳)有限公司 一种基于点云扫描的数控机床工件加工质量控制方法
CN115145221B (zh) * 2022-07-05 2024-07-19 湖南大学 一种基于机器人铣边误差追溯的工件与刀具位姿标定方法
CN115408785B (zh) * 2022-07-12 2026-04-07 上海拓璞数控科技股份有限公司 三维模型的自适应补偿方法及系统
CN115369223B (zh) * 2022-07-28 2023-11-21 华东理工大学 一种双侧超声滚压协同强化系统及其控制方法
CN115507803B (zh) * 2022-10-20 2025-02-07 中国航发北京航空材料研究院 一种涡轮叶片加工和检测的方法、装置及设备
CN116839499B (zh) * 2022-11-03 2024-04-30 上海点莘技术有限公司 一种大视野微尺寸2d及3d测量标定方法
CN116244790B (zh) * 2022-12-22 2026-03-13 上海建工四建集团有限公司 系杆拱桥钢拱肋加工质量评估方法
CN116165340B (zh) * 2022-12-28 2023-09-29 杭州中为光电技术有限公司 一种硅棒在线检测系统及检测方法
CN115931908B (zh) * 2022-12-28 2024-05-07 杭州中为光电技术有限公司 一种硅棒缺陷自动检测系统及检测方法
CN116466649B (zh) * 2023-04-25 2025-09-19 浙江大学 一种基于三维激光扫描分析的机床加工系统
CN116476387A (zh) * 2023-05-26 2023-07-25 深圳市创想三维科技股份有限公司 基于3d打印的补偿方法、电子设备及计算机可读存储介质
CN116403208B (zh) * 2023-06-07 2023-08-22 山东科技大学 基于激光雷达点云的滚轮罐耳运行状态检测方法及装置
CN117124483B (zh) * 2023-07-13 2024-03-08 同济大学 基于在线与离线检测的自由曲面棱镜高精度补偿加工方法
CN117055467B (zh) * 2023-09-13 2024-09-06 安徽久泰电气有限公司 一种五轴加工机器人自动化控制系统及控制方法
CN117245151B (zh) * 2023-09-28 2025-09-19 中南大学 一种超硬齿轮超声辅助高效精密磨齿加工装置与方法
CN117245583B (zh) * 2023-11-01 2025-10-28 沈阳飞机工业（集团）有限公司 精准可调定位工装及定位方法
CN117840492A (zh) * 2023-11-30 2024-04-09 哈尔滨汽轮机厂有限责任公司 一种自适应补偿叶片根冠径向面平行度的数控加工方法
CN118111350A (zh) * 2024-01-23 2024-05-31 哈尔滨工业大学 基于结构光双向同步扫描的航空发动机叶片型面测量装置与方法
CN118466387A (zh) * 2024-04-24 2024-08-09 上海拓璞数控科技股份有限公司 镜像铣加工与测控工艺方法及其系统
CN119085526B (zh) * 2024-08-29 2025-09-09 北京航空航天大学 一种桨叶型面精度测量系统及其误差分析方法
CN120593657A (zh) * 2025-08-07 2025-09-05 御传(上海)传动科技有限公司 一种减速电机针轮齿形误差的激光扫描检测方法
CN121048554A (zh) * 2025-09-18 2025-12-02 浙江佳太汽车零部件有限公司 一种汽车空调皮带轮的圆跳动质检系统及检测装置

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100079769A1 (en) * 2008-09-29 2010-04-01 General Electric Company Method and system for parameter extraction of a cutting tool
CN103630085A (zh) * 2013-10-15 2014-03-12 广东工业大学 多角度叶片测量装置及方法
CN104392488A (zh) * 2014-12-11 2015-03-04 福州大学 针对激光扫描仪与三坐标测量臂的点云数据自动配准方法
CN105157609A (zh) * 2015-09-01 2015-12-16 大连理工大学 基于两组相机的大型零件全局形貌测量方法
WO2016020175A1 (de) * 2014-08-07 2016-02-11 Walter Maschinenbau Gmbh Verfahren und vorrichtung zur bestimmung wenigstens eines modellparameters eines virtuellen werkzeugmodells eines werkzeugs
CN105737762A (zh) * 2016-05-09 2016-07-06 中国民用航空飞行学院 一种航空发动机叶片型面测量方法
CN107246849A (zh) * 2017-05-25 2017-10-13 西安知象光电科技有限公司 一种基于双测头四轴测量系统的叶片光学快速测量方法
CN109489580A (zh) * 2018-12-10 2019-03-19 华东理工大学 一种复杂表面加工的在机点云检测及补偿方法

Family Cites Families (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4838696A (en) * 1983-01-28 1989-06-13 Diffracto Ltd. Pulsed robotic inspection
JP4897572B2 (ja) * 2006-06-30 2012-03-14 株式会社ミツトヨ 斜入射干渉計
CN101853485A (zh) * 2010-06-04 2010-10-06 浙江工业大学 一种基于近邻传播聚类的非均匀点云简化处理方法
US9367770B2 (en) * 2011-08-30 2016-06-14 Digimarc Corporation Methods and arrangements for identifying objects
US10203368B2 (en) * 2012-01-06 2019-02-12 New York University Systems, processes and computer-accessible medium for providing a bi-directional scan path for peak capture power reduction in launch-off-shift testing
CN105513127B (zh) * 2015-12-25 2018-01-26 武汉大学 基于密度峰值聚类的杆状物规则化三维建模方法及系统
CN205363482U (zh) * 2016-01-22 2016-07-06 浙江优傲智能科技有限公司 自动化打磨抛光集成系统
CN205870292U (zh) * 2016-06-27 2017-01-11 浙江电驱机器人有限公司 一种自动抛光打磨装置
US10395372B2 (en) * 2016-06-28 2019-08-27 University Of Cincinnati Systems, media, and methods for pre-processing and post-processing in additive manufacturing
CN106643551B (zh) * 2016-12-07 2019-07-30 苏州笛卡测试技术有限公司 一种叶片形状快速扫描装置及方法
CN107238352B (zh) * 2017-06-01 2019-07-09 南京航空航天大学 一种基于数控车床的回转类结构特征零件轮廓激光在机测量装置与方法
CN108330264A (zh) * 2017-12-22 2018-07-27 华东理工大学 一种轨迹可控的表面强化装置
CN108038906B (zh) * 2017-12-26 2021-04-02 山东师范大学 一种基于图像的三维四边形网格模型重建方法
CN108458659A (zh) * 2018-05-03 2018-08-28 华中科技大学无锡研究院 一种叶片非接触式检测装置及方法
CN108830931B (zh) * 2018-05-23 2022-07-01 上海电力学院 一种基于动态网格k邻域搜索的激光点云精简方法
CN108765478A (zh) * 2018-06-06 2018-11-06 杭州中科天维科技有限公司 一种建筑点云中分离单体建筑的密度聚类算法

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100079769A1 (en) * 2008-09-29 2010-04-01 General Electric Company Method and system for parameter extraction of a cutting tool
CN103630085A (zh) * 2013-10-15 2014-03-12 广东工业大学 多角度叶片测量装置及方法
WO2016020175A1 (de) * 2014-08-07 2016-02-11 Walter Maschinenbau Gmbh Verfahren und vorrichtung zur bestimmung wenigstens eines modellparameters eines virtuellen werkzeugmodells eines werkzeugs
CN104392488A (zh) * 2014-12-11 2015-03-04 福州大学 针对激光扫描仪与三坐标测量臂的点云数据自动配准方法
CN105157609A (zh) * 2015-09-01 2015-12-16 大连理工大学 基于两组相机的大型零件全局形貌测量方法
CN105737762A (zh) * 2016-05-09 2016-07-06 中国民用航空飞行学院 一种航空发动机叶片型面测量方法
CN107246849A (zh) * 2017-05-25 2017-10-13 西安知象光电科技有限公司 一种基于双测头四轴测量系统的叶片光学快速测量方法
CN109489580A (zh) * 2018-12-10 2019-03-19 华东理工大学 一种复杂表面加工的在机点云检测及补偿方法

Cited By (35)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111798570B (zh) * 2020-07-16 2022-06-17 浙江大学 一种叶轮机械叶片曲面网格化快速生成方法
CN111798570A (zh) * 2020-07-16 2020-10-20 浙江大学 一种叶轮机械叶片曲面网格化快速生成方法
CN112215841A (zh) * 2020-10-31 2021-01-12 上海扩博智能技术有限公司 风机叶片点云数据的形成方法、系统、设备和存储介质
CN112731864A (zh) * 2020-12-21 2021-04-30 北京理工大学 一种机床加工误差补偿方法、装置及零件加工机床
CN112581579A (zh) * 2020-12-22 2021-03-30 同济大学 一种磁浮滑行面点云数据提取方法
CN112581579B (zh) * 2020-12-22 2022-11-18 同济大学 一种磁浮滑行面点云数据提取方法
CN112626321A (zh) * 2020-12-28 2021-04-09 华南理工大学 Rv减速器摆线轮超声滚压过程中的变形控制装置及方法
CN112626321B (zh) * 2020-12-28 2024-04-05 华南理工大学 Rv减速器摆线轮超声滚压过程中的变形控制装置及方法
CN112762856A (zh) * 2020-12-29 2021-05-07 中国航空工业集团公司西安飞机设计研究所 一种用于扫描影响因素的评价方法
CN112762856B (zh) * 2020-12-29 2023-10-20 中国航空工业集团公司西安飞机设计研究所 一种用于扫描影响因素的评价方法
CN113390340A (zh) * 2021-06-15 2021-09-14 中国工程物理研究院机械制造工艺研究所 一种非连续区域的角球面球心空间位置检测方法
CN113610916A (zh) * 2021-06-17 2021-11-05 同济大学 基于点云数据的不规则物体体积测定方法及系统
CN113610916B (zh) * 2021-06-17 2024-04-12 同济大学 基于点云数据的不规则物体体积测定方法及系统
CN113369998A (zh) * 2021-06-30 2021-09-10 中国航发动力股份有限公司 一种基于工艺模型的模锻叶片型面自适应补偿加工方法
CN113985813B (zh) * 2021-10-27 2023-09-01 中国航发沈阳黎明航空发动机有限责任公司 一种基于在机检测的机床原点误差补偿方法
CN113985813A (zh) * 2021-10-27 2022-01-28 中国航发沈阳黎明航空发动机有限责任公司 一种基于在机检测的机床原点误差补偿方法
CN114187402A (zh) * 2021-11-04 2022-03-15 河南农业大学 基于三维点云的烤烟三维模型构建方法
CN114393448B (zh) * 2022-01-21 2023-11-28 中国科学院长春光学精密机械与物理研究所 一种提高磁流变机器人抛光设备轨迹精度的方法
CN114393448A (zh) * 2022-01-21 2022-04-26 中国科学院长春光学精密机械与物理研究所 一种提高磁流变机器人抛光设备轨迹精度的方法
CN114777670A (zh) * 2022-04-21 2022-07-22 西安交通大学 一种基于接触式测头的曲面在机测量方法
CN115647438A (zh) * 2022-10-31 2023-01-31 首都航天机械有限公司 一种超大径厚比整体箱底自适应补偿加工的找正方法
CN116372668A (zh) * 2023-03-10 2023-07-04 珠海格力精密模具有限公司 一种工件加工精度检测修正方法、检测系统及电子设备
CN116512806A (zh) * 2023-04-28 2023-08-01 华侨大学 面向石雕切割工件测量的机器人扫描视点规划方法及系统
CN116952181B (zh) * 2023-09-18 2024-01-12 成都飞机工业（集团）有限责任公司 一种大长径比复合材料管型薄壁件的内型面测量方法
CN116952181A (zh) * 2023-09-18 2023-10-27 成都飞机工业（集团）有限责任公司 一种大长径比复合材料管型薄壁件的内型面测量方法
CN117806231A (zh) * 2024-02-27 2024-04-02 山东微晶重工有限公司 一种基于物联网的机床运行加工控制系统及方法
CN117806231B (zh) * 2024-02-27 2024-05-03 山东微晶重工有限公司 一种基于物联网的机床运行加工控制系统及方法
CN118623797A (zh) * 2024-05-31 2024-09-10 天津大学 一种多层结构三维信息的测量方法及装置
CN118513913A (zh) * 2024-07-19 2024-08-20 无锡康贝电子设备有限公司 一种数控机床不规则零件的尺寸测量方法
CN118513913B (zh) * 2024-07-19 2024-11-15 无锡康贝电子设备有限公司 一种数控机床不规则零件的尺寸测量方法
CN119115564A (zh) * 2024-10-16 2024-12-13 吉林大学 激光加热超声振动辅助超精密加工车铣一体机床及控制方法
CN119251292A (zh) * 2024-12-05 2025-01-03 福建省武夷山市永生茶业有限公司 用于茶陇采摘的割刀定位方法、装置、电子设备及存储介质
CN119984095A (zh) * 2025-04-11 2025-05-13 远东幕墙(珠海)有限公司 一种双曲铝板加工用测量装置
CN120235876A (zh) * 2025-05-30 2025-07-01 西安爱德华测量设备股份有限公司 一种精锻叶片进排气边双缺口叶型轮廓度检测方法
CN120791523A (zh) * 2025-08-27 2025-10-17 天津蕴宝伟业科技发展有限公司 一种无尘打磨自动生产线的板材加工控制方法及系统

Also Published As

Publication number Publication date
CN109489580A (zh) 2019-03-19
US20220057194A1 (en) 2022-02-24
US12061078B2 (en) 2024-08-13
CN109489580B (zh) 2021-09-24

Similar Documents

Publication Publication Date Title
WO2020118970A1 (zh) 一种复杂表面加工的在机点云检测及补偿方法
CN104515478B (zh) 一种高精度的航空发动机叶片自动三维测量方法和系统
CN103433810B (zh) 一种复杂曲面法矢在机检测装置及方法
CN103777570B (zh) 基于nurbs曲面的加工误差快速检测补偿方法
Ding et al. Error modeling and path planning for freeform surfaces by laser triangulation on-machine measurement
Ding et al. Evaluation and compensation of laser-based on-machine measurement for inclined and curved profiles
CN115112018A (zh) 一种基于立体视觉的三坐标机智能复合联检系统及方法
CN113865508A (zh) 一种蜂窝夹芯复合材料声衬的通孔率自动化检测装置与方法
Chang et al. Automatic inspection of turbine blades using a 3-axis CMM together with a 2-axis dividing head
CN204269086U (zh) 一种高精度的航空发动机叶片自动三维测量系统
Tian et al. A technology framework for robotic profiling of blade edges based on model reconstruction and trajectory replanning
CN116466649A (zh) 一种基于三维激光扫描分析的机床加工系统
KR101622659B1 (ko) 로봇과 레이저 비전 시스템 간의 캘리브레이션 방법
Wang et al. On-machine volumetric-error measurement and compensation methods for micro machine tools
CN110375674A (zh) 一种精密制造设备的视觉检测系统
CN108662989A (zh) 一种基于三维激光扫描的车灯轮廓质量检测方法
Shen et al. Measurement and evaluation of laser-scanned 3D profiles in wire arc hybrid manufacturing processes
CN120800298A (zh) 一种非接触式零件圆弧参数高精度测量方法及系统
Kou et al. Calibration of the laser displacement sensor and integration of on-site scanned points
Sun et al. On-machine measurement system for roundness and axis straightness errors in deep hole boring
CN108115466B (zh) 一种立式加工中心的几何运动误差检测方法及系统
CN110405259A (zh) 基于多传感器集成测量的自由曲面类零件加工系统
CN116086322B (zh) 针对高反光碗形冲压件成形尺寸自动化检测方法和系统
Liu et al. Rotation scanning and double precision system for three-dimensional geometric on-machine measurement
Tong et al. A novel laser-based system for measuring internal thread parameters

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19896361

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19896361

Country of ref document: EP

Kind code of ref document: A1