Multi-size and form and position error comprehensive measurement method and system for large metal disc parts

Abstract

The invention discloses a multi-size and form and position error comprehensive measurement method and system for a large metal disc part, and belongs to the technical field of precision measurement. According to the method, a workpiece is clamped on a high-precision rotating table, and three-dimensional point cloud data of the complete surface of the workpiece are obtained through cooperative movement of a non-contact three-dimensional scanning measuring head and the high-precision rotating table; unifying all the point clouds into a global coordinate system taking the rotation axis as a reference; geometric elements are fitted under the coordinate system, and sizes such as diameter and thickness and various form and position errors such as coaxiality, perpendicularity, parallelism and planeness are calculated at a time. The corresponding system comprises a rotary table, a three-dimensional mobile platform, a scanning probe, a control system and a data processing unit. According to the invention, multi-parameter, high-precision and high-efficiency automatic measurement of large disc parts is realized, the reference is unified, multiple clamping errors are avoided, and the method is especially suitable for precision detection in the field of high-end equipment manufacturing.

Description

Technical Field

[0001] This invention relates to precision measurement technology, specifically to a method and system for comprehensive measurement of multiple dimensions and form and position errors of large metal disc-shaped parts. Background Technology

[0002] Large metal disc-shaped parts are key basic components in critical equipment (such as aero engines, gas turbines, and wind turbine generators), and their dimensional accuracy and geometric tolerances directly affect the performance, reliability, and lifespan of the entire machine. Traditional measurement methods mainly rely on tools such as coordinate measuring machines (CMMs), large micrometers, dial indicators, and levels. However, existing technologies have the following significant drawbacks:

[0003] Inefficient: For the measurement of multiple dimensions and multiple geometric errors, multiple clamping, changing different tools, and performing the measurement on different equipment are required. The measurement process is cumbersome and time-consuming.

[0004] Risks of contact measurement: Contact measurement methods such as CMM may scratch the surface of the part, and the measurement force may introduce errors, which is especially disadvantageous for thin-walled or easily deformable large disc-shaped parts.

[0005] Systematic errors: Transferring and repeatedly clamping parts between different devices can introduce positioning errors, making it difficult to ensure that all measurement results are based on a unified datum, thus affecting the accuracy of geometric tolerance evaluations such as coaxiality and perpendicularity that require a common datum.

[0006] Low level of automation: The system heavily relies on operator experience; manual reading, recording, and data processing are prone to errors and cannot achieve rapid, online inspection on the production floor. Therefore, there is an urgent need for an automated measurement solution capable of achieving multi-parameter, high-precision, high-efficiency, non-contact measurement of large disc-shaped parts with a unified reference standard. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method and system for comprehensive measurement of multiple dimensions and form and position errors of large metal disc-shaped parts. To achieve the above objective, this invention adopts the following technical solution.

[0008] This invention provides a comprehensive measurement method for multiple dimensions and form and position errors of large metal disc-shaped parts, comprising the following steps:

[0009] S1: Datum Establishment and Workpiece Clamping. Place the large metal disc-shaped part to be measured on a high-precision rotary table. Using a laser tracker or vision positioning system, precisely adjust the position of the workpiece so that its theoretical axis initially coincides with the rotation axis of the rotary table, thus establishing the measurement datum.

[0010] S2: 3D Point Cloud Data Acquisition. Activate the non-contact 3D scanning probe (such as a laser line scanning probe or structured light scanning probe) fixed on the 3D moving platform. Control the coordinated movement of the 3D moving platform and the rotary table, causing the scanning probe to scan the entire measured surface of the part (including end faces, inner and outer cylindrical surfaces, stepped surfaces, etc.) along a preset path, simultaneously acquiring high-density 3D point cloud data. Simultaneously, record the angle encoder data of the rotary table, correlating the point cloud data with the rotation angle.

[0011] S3: Point Cloud Preprocessing and Coordinate System I. The massive amount of point cloud data collected is denoised, filtered, and simplified. Using the reference established in step S1 and the rotation center of the rotary table, all point clouds collected at different angles are unified into a global coordinate system with the rotary table axis as the Z-axis.

[0012] S4: Digital fitting of key geometric features. Based on a unified global coordinate system point cloud, key geometric features are digitally fitted using algorithms such as least squares and RANSAC.

[0013] Fit the outer cylindrical surface / inner cylindrical surface to obtain its axis L1 and diameter D1.

[0014] Fit the main end face to obtain its plane P1.

[0015] By fitting other feature surfaces (such as the end face of the step, the axis of the auxiliary positioning hole, etc.), plane P2, axis L2, etc. are obtained.

[0016] S5: Calculation of multiple dimension parameters. Based on the geometric features fitted in step S4, the required dimension parameters are directly calculated, for example:

[0017] Diameter: Obtained directly from the fitted cylindrical surface.

[0018] Thickness dimension: obtained by calculating the distance between the two end faces (P1 and P2) in the axial direction.

[0019] Groove width, step height, etc. S6: Geometric shape and position error evaluation. Under a unified global coordinate system, based on the fitted geometric elements and the references required by the drawing, calculate the geometric shape and position errors:

[0020] Coaxiality error: Using the reference axis (such as the inner hole axis L1) as a reference, evaluate twice the maximum offset of the measured axis (such as the outer circle axis or another step axis) relative to the reference axis.

[0021] Perpendicularity error: Using the reference axis (such as L1) as a reference, evaluate the maximum distance from all points on the measured end face (P1) to the ideal plane perpendicular to the reference axis.

[0022] Parallelism error: Using a reference plane (such as P1) as a reference, evaluate the maximum distance from all points on the measured plane (P2) to an ideal plane parallel to the reference plane.

[0023] Flatness error: For the measured end face (P1) itself, it is the sum of the absolute values ​​of the maximum positive deviation and the maximum negative deviation of its point cloud data from the fitted ideal plane.

[0024] S7: Results Output and Visualization. Generates an inspection report containing all measured dimensions and geometrical error values, and visualizes the error distribution on a 3D point cloud model as a color cloud map or deviation vector.

[0025] This invention provides a comprehensive measurement system for multiple dimensions and form and position errors of large metal disc-shaped parts, used to implement the above method, specifically including:

[0026] High-precision rotary table: used to support and precisely drive the rotation of the part to be measured, with its rotation axis serving as the reference axis for measurement.

[0027] Three-dimensional moving platform: usually a gantry type or high-precision three-axis guide rail structure, providing precise movement in the X, Y and Z directions.

[0028] Non-contact 3D scanning probe: fixed on a 3D moving platform, used to quickly acquire 3D point cloud data of the surface of a part.

[0029] Laser tracker or vision positioning system (optional but preferred): used for initial workpiece positioning and global system calibration to ensure consistent reference.

[0030] Control system: It communicates with the rotary table, three-dimensional moving platform, scanning probe and laser tracker to control them to move in coordination along a predetermined trajectory.

[0031] Data processing and evaluation unit: A computer with dedicated software, used to receive point cloud data, execute the aforementioned point cloud preprocessing, coordinate system one, geometric element fitting, size calculation and form and position error evaluation algorithms, and output the final results.

[0032] The beneficial effects of this invention are:

[0033] High efficiency and automation: All parameters to be measured can be automatically measured in a single clamping, which greatly shortens the detection time and improves the detection efficiency.

[0034] High precision and completeness: Non-contact measurement avoids errors caused by measurement force, and high-density point clouds can fully reflect the micro-geometric features of the part surface, resulting in more comprehensive and accurate evaluation results.

[0035] Unified datum: All measurements are completed in a unified coordinate system with the axis of the rotary table as the core, which fundamentally eliminates the errors caused by multiple clamping and datum conversion, and ensures the accuracy of evaluation of geometric tolerances such as coaxiality and perpendicularity.

[0036] Digitalization and visualization: Generates digital inspection reports and 3D deviation maps, facilitating quality traceability, process analysis, and product improvement. Attached Figure Description

[0037] Figure 1 This is a schematic diagram of the measurement system of the present invention.

[0038] Figure 2 This is a schematic diagram of the rotary table structure of the present invention.

[0039] Figure 3 This is a schematic diagram of the non-contact three-dimensional scanning probe structure of the present invention.

[0040] Figure 4 This is a flowchart of the measurement method of the present invention.

[0041] Figure 5 This is a schematic diagram of point cloud data and fitted geometric elements.

[0042] Figure 6 This represents the distance error distribution from the point cloud to the fitted plane.

[0043] The components are: 1-frame; 2-non-contact 3D scanning probe; 3-large metal disc-like parts; 4-rotary worktable; 5-base. Detailed Implementation

[0044] This section describes in detail a preferred embodiment of the present invention, enabling those skilled in the art to implement the invention. (See attached drawing.) Figure 1-6 The following description covers the system hardware configuration, control logic, software algorithm, and verification cases. The implementation method uses a large metal disc-shaped part as an example, but this invention is applicable to various metal disc-shaped parts, such as gear discs and flanges.

[0045] Example 1: The core structure and assembly relationship of this measuring device are as follows Figure 1As shown, frame 1 serves as the overall supporting skeleton, providing the installation reference and spatial positioning for each component. The non-contact 3D scanning probe 2 is rigidly fixed to the pre-set mounting position on the frame. The installation height and axial angle are precisely designed based on the scanning angle to ensure the scanning coverage. Base 5 serves as the base for the rotating unit, and the rotary table 4 is fixed with bolts. Large metal disc-shaped parts 3 are precisely clamped onto the worktable surface of the rotary table 4 using a customized 3D-printed fixture. After clamping, laser alignment fixtures or mechanical limit blocks are used to ensure that the measured feature area of ​​the part is completely aligned with the scanning field of view of the non-contact 3D scanning probe 2, thus providing hardware-level assurance for the integrity and measurement accuracy of subsequent scanning data.

[0046] Example 2: System Hardware Composition and Initialization Configuration. The measurement system of the present invention is as follows: Figure 1 As shown, it mainly includes the following core components. Each component is synchronously controlled through the EtherCAT bus, with a communication cycle of ≤1ms.

[0047] 1. High-precision rotary table:

[0048] Mechanical structure: Supported by air bearings, radial runout ≤0.8μm, axial runout ≤0.3μm; drive system is a torque motor direct drive, equipped with a 17-bit absolute encoder, speed range 0.00120rpm, speed regulation resolution 0.0001rpm.

[0049] Initial setup: The worktable is equipped with a hydraulic clamp, supporting a maximum workpiece weight of 500kg and a repeatability of clamping accuracy ≤0.01mm. The rotation axis is calibrated using a laser tracker and used as the Z-axis reference of the global coordinate system.

[0050] 2. Three-dimensional mobile platform:

[0051] The preferred structure is a gantry type, with a granite base. The XYZ three-axis travel is 5000mm×3000mm×1500mm. It is driven by a linear motor, with a positioning accuracy of ±0.005mm and a repeatability of ±0.002mm.

[0052] The mobile platform is equipped with a non-contact scanning probe and integrates an anti-collision sensor.

[0053] 3. Non-contact 3D scanning probe:

[0054] Model: Blue laser line scanner, 6400 points / mm² per line, sampling frequency 500Hz, measurement accuracy ±0.025mm; or structured light scanning probe, resolution 2448×2048 pixels.

[0055] Adaptive adjustment: The probe holder is equipped with an electric tilt mechanism, which can dynamically adjust the posture according to the curvature of the workpiece, and dynamically adjust the laser power to avoid overexposure due to reflection.

[0056] 4. Control system and data processing unit:

[0057] An industrial computer is installed with dedicated measurement software; the control system is based on a PLC and a G-code parser to realize multi-axis cooperative motion planning.

[0058] The laser tracker is used for initial positioning of the workpiece. The target ball is arranged according to the "321" principle to fit the theoretical axis of the workpiece to coincide with the axis of rotation.

[0059] Example 3: Detailed steps of the measurement process, corresponding to steps S1-S7 of the invention, are implemented as follows:

[0060] Step S1: Adjust the position of the large metal disc-shaped part to be measured and establish a measurement reference.

[0061] Step S2: Synchronously acquire high-density 3D point cloud data and rotary table angle encoder data, and correlate the two.

[0062] Step S3: Denoise, filter and simplify the massive point cloud data collected.

[0063] Step S4: Based on the unified global coordinate system point cloud, the key geometric elements are digitally fitted.

[0064] Step S5: Based on the fitted geometric features, directly calculate the required dimensional parameters.

[0065] Step S6: Calculate the form and position errors based on the fitted geometric elements and the references required by the drawing.

[0066] Step S7: Generate an inspection report and visualize the error distribution.

[0067] In step S1, the specific steps are as follows:

[0068] S11: The turbine disk to be tested is clamped on the rotary table, and three feature points—the end face of the disk and the inner hole—are captured by the laser tracker to establish the local coordinate system of the workpiece.

[0069] S12: Use the ICP (Iterative Closest Point) algorithm to align the local coordinate system with the rotation axis, and adjust the workpiece position until the axis deviation is ≤0.01mm. Record the initial pose matrix as the measurement reference.

[0070] In step S2, the specific steps are as follows:

[0071] S21: Control Strategy: The rotary table rotates in indexing mode, pausing every 1.5°. The three-dimensional moving platform drives the scanning probe to perform a serpentine scan along the workpiece's axial (Z-axis) and radial (X-axis) directions. The scanning speed is matched to the rotational speed (V_scan=ω×R) to avoid motion ambiguity.

[0072] S22: Data Synchronization: The angle encoder records corner data in real time, and the point cloud data is associated with the corner and stored in a time-series format. A single scan covers the entire surface, with a point cloud density ≥10,000 points / cm².

[0073] In step S3, the specific steps are as follows:

[0074] S31: Denoising: Statistical outlier filtering (k=50, standard deviation threshold 2.0) is used to remove drift points.

[0075] S32: Registration: Multi-view point cloud stitching based on KDtree, using ICP algorithm to optimize stitching error (RMS≤0.005mm).

[0076] S33: Coordinate Unification: Establish a global coordinate system with the rotation axis as the Z-axis and the center of the worktable as the origin. All point clouds are transformed to this coordinate system through a rigid body transformation matrix.

[0077] In step S4, the specific steps are as follows:

[0078] S41: Cylindrical Surface Fitting: For the point clouds of the inner and outer cylindrical surfaces, the least squares method is used to fit the axis L1 and the diameter D1; the fitting residual RMS ≤ 0.01 mm.

[0079] S42: Plane Fitting: For the end face point cloud, use the RANSAC algorithm to fit planes P1 and P2 and remove outliers.

[0080] S43: Other features: Step surfaces, hole axes, etc. are fitted using similar methods to obtain a geometric element set {L1,D1,P1,P2,L2}.

[0081] In step S5, the specific steps are as follows:

[0082] S51: Diameter dimension: Directly derived from the fitted cylindrical surface equation, such as outer circle diameter D_outer = 2 × cylinder radius.

[0083] S52: Thickness dimension: Calculate the average distance between the two end faces P1 and P2 in the axial direction, formula: Thickness=|P1.center P2.center|·n_axis (n_axis is the unit vector of the axis).

[0084] S53: Slot width / step height: Calculated using the Euclidean distance between feature point clouds.

[0085] In step S6, the specific steps are as follows:

[0086] S61: Coaxiality error: Taking the inner hole axis L1 as the reference, calculate the radial distance from each point on the outer circle axis L2 to L1, and take twice the maximum distance: Coaxiality=2×max(||L2_point L1_projection||).

[0087] S62: Verticality error: Based on L1, construct an ideal vertical plane and calculate the maximum distance from the point cloud of end face P1 to this plane: Perpendicularity=max(|(P P0)·n_vertical|).

[0088] S63: Parallelism error: Using P1 as the reference, calculate the maximum distance from the P2 point cloud to the ideal parallel plane.

[0089] S64: Flatness error: For point cloud P1, calculate the sum of the absolute values ​​of the maximum positive and negative deviations to the fitted ideal plane: Flatness=|max(δ_positive)|+|min(δ_negative)|.

[0090] In step S7, the specific steps are as follows:

[0091] S71: Visualization: Renders color deviation cloud maps (e.g., flatness error represented by red-blue gradient) on a 3D point cloud model, supporting section views and vector annotations (e.g. Figure 4 (As shown).

[0092] Example 4: Key Algorithms and Software Implementation: Core Code Example for Point Cloud Processing (Based on PCL Library):

[0093] / / Pseudocode example: Plane fitting and error calculation

[0094] pcl::PointCloud <pcl::pointxyz>::Ptr cloud(new pcl::PointCloud <pcl::pointxyz>);

[0095] pcl::SACSegmentation <pcl::pointxyz>seg;

[0096] seg.setModelType(pcl::SACMODEL_PLANE);

[0097] seg.setMethodType(pcl::SAC_RANSAC);

[0098] seg.setDistanceThreshold(0.01);

[0099] pcl::ModelCoefficients coefficients;

[0100] seg.segment(*inliers,coefficients);

[0101] / / Calculate flatness error

[0102] float flatness_error=computeFlatness(cloud,coefficients);

[0103] Control logic: Coordinated motion is planned using a trapezoidal velocity curve to avoid sudden stop vibration; data transmission is carried out through the TCP / IP protocol to ensure real-time performance.

[0104] Example 5: Performance Verification and Implementation Case A case study using a gas turbine turbine disk as an example:

[0105] Measurement results: Parameter measured values ​​tolerance requirements and measurement uncertainty: Outer diameter Φ120 0.005mm ±0.02mm ±0.003mm, end face flatness 0.008mm ≤0.015mm ±0.001mm, inner hole coaxiality 0.006mm ≤0.01mm ±0.002mm.

[0106] Efficiency improvement: Single-piece measurement time ≤ 10 minutes (traditional CMM requires 2 hours), repeatability accuracy ≤ 0.005mm (k=2).

[0107] The above-described embodiments, through hardware selection, algorithm refinement, and case verification, demonstrate the high precision and high efficiency advantages of this invention. Those skilled in the art can adjust the parameters based on the above description to adapt to different workpiece sizes, all of which fall within the protection scope of this invention.< / pcl::pointxyz> < / pcl::pointxyz> < / pcl::pointxyz>

Claims

1. A comprehensive measurement method for multiple dimensions and form and position errors of large metal disc-shaped parts, characterized in that, Includes the following steps: The workpiece is clamped on a high-precision rotary table and a measurement datum is established. Control the non-contact 3D scanning probe to move in coordination with the rotary table to collect 3D point cloud data of the complete surface of the workpiece; Unify all point cloud data into a global coordinate system based on the rotation axis of the rotary table; In the global coordinate system, the point cloud data is processed to fit the key geometric elements of the workpiece. Based on the fitted geometric features, multiple dimensional parameters of the workpiece are calculated; In the global coordinate system, based on the measurement datum and the fitted geometric elements, at least two of the form and position errors among the coaxiality, perpendicularity, parallelism and flatness of the workpiece are calculated.

2. The method according to claim 1, characterized in that, The step of establishing the measurement benchmark includes using a laser tracker or vision positioning system to adjust the workpiece so that its theoretical axis coincides with the rotation axis of the rotary table.

3. The method according to claim 1, characterized in that, The coordinated motion refers to the following: the rotary table rotates in increments, and after pausing at each rotation angle, the three-dimensional moving platform drives the scanning probe to perform linear scanning along the workpiece axial or radial direction.

4. The method according to claim 1, characterized in that, The calculation of the form and position error includes: Coaxiality error is calculated by taking the fitted reference axis as the reference and calculating the maximum radial offset of the measured axis from the reference axis. Perpendicularity error is calculated by taking the fitted reference axis as a reference and calculating the maximum distance between the measured surface and the ideal plane perpendicular to the reference axis. Parallelism error is calculated by taking the fitted reference plane as a reference and calculating the maximum distance between the measured surface and an ideal plane that is parallel to the reference plane. Flatness error is calculated as the sum of the absolute values ​​of the maximum positive and negative deviations of the point cloud data of the measured surface from its fitted ideal plane.

5. A comprehensive measurement system for multiple dimensions and form and position errors of large metal disc-shaped parts, used to implement the method described in any one of claims 1 to 4, characterized in that, include: A high-precision rotary table is used to carry and precisely drive the rotation of the part to be measured, with its axis of rotation serving as the reference axis for measurement. A mobile platform capable of three-dimensional movement, providing precise motion in the X, Y, and Z directions; A non-contact 3D scanning probe mounted on the mobile platform; A control system, based on a PLC and a G-code parser, communicates with a rotary table, a three-dimensional moving platform, a scanning probe, and a laser tracker to achieve multi-axis collaborative motion planning. A data processing and evaluation unit is used to receive point cloud data from the scanning probe and perform point cloud processing, geometric feature fitting, size calculation and form and position error evaluation.

6. The system according to claim 5, characterized in that, It also includes a laser tracker or vision positioning system for initial workpiece positioning and system calibration.

7. The system according to claim 5, characterized in that, The non-contact three-dimensional scanning probe is either a laser line scanning probe or a structured light scanning probe.

8. The system according to claim 5, characterized in that, The mobile platform has a gantry structure.

Images