| Blade is an important part of aero-engine,gas turbine and other power machinery.The machining quality of its surface seriously affects the working performance of the whole machine,so it is very necessary to detect its surface profile.The blade surface is generally a torsion surface with variable cross-section,and the curvature of the blade surface changes greatly,so it is difficult to detect.The traditional three coordinate measurement method evaluates the machining quality of blade by obtaining the profile data of specific section of blade surface,which is suitable for the detection of finishing blade,and can not reflect the overall size deviation of blade.At the same time,it has high requirements for the detection environment,and can not realize the on-line detection of blade.With the development of optical measurement technology,line structured light method has the advantages of fast measurement speed,high accuracy and non-contact.It can quickly obtain the profile data of the blade profile to be measured,which provides a feasible technical means for the high-precision and efficient detection of the blade profile.Based on the made in China 2025 Sichuan Action Fund Project(No.: 2018CD00225),combined with the actual needs of enterprises to establish blade intelligent remanufacturing factory,this paper studies the blade profile detection method based on line structured light.It involves the calibration method of the position between the linear structured light sensor and the blade to be measured,the establishment of the comprehensive error model of the blade detection system based on the multi-body system theory,the measurement error analysis of the linear structured light sensor,the data acquisition of the blade profile based on the standard cylinder and the three-dimensional shape reconstruction.The main research contents and achievements of this paper are as follows:(1)Based on the principle of line structured light measurement,a set of blade surface detection platform is designed.The platform uses three linear axes and turntable to realize the relative motion between the measuring device and the blade to be measured,so that the measuring device can complete the multi view data acquisition of the blade,and finally complete the selection of the platform moving parts and the line structured light sensor.(2)The calibration method of the linear structured light sensor and the blade to be measured is studied,so that the relationship between them is correct,the detection accuracy is improved,and the results meet the universality of the detection.By studying the influence of the installation posture of the line structured light sensor on the measurement results,the inclination sensor and calibration block are introduced to complete the calibration of the line structured light sensor posture;the machining datum plane at the tenon is used to calibrate the blade axis to achieve the blade posture calibration.The calibration method reduces the influence of the installation error of the linear structured light sensor and the blade to be measured on the measurement results,and makes the measurement results have the standard of comparison.(3)The main error sources of the blade detection platform are studied,and 27 geometric errors of the motion system are analyzed.Based on the multi-body system theory,the topological structure of the detection system is established,and the transformation matrix between the theoretical and actual motion of each component of the system is solved by using the homogeneous coordinate transformation method,and the comprehensive error model of the detection system is established;The geometric error identification of the motion system is completed by using the laser tracker,which provides data support for the error compensation of the high-precision motion control of the detection system;The static measurement error of linear structured light sensor in different measurement distance,incidence angle and surface roughness is explored by using a variety of roughness comparison blocks,which lays a foundation for reducing the measurement error of linear structured light sensor and improving the accuracy of blade detection.(4)The measurement and evaluation methods of blade profile are studied.By introducing the standard cylinder and using its center coordinates to realize the data stitching of cross-section profile;By NURBS fitting and equidistant sampling processing of point cloud data,the registration of measurement point cloud and theoretical point cloud is realized;According to the definition of profile degree and main characteristic parameters,the blade detection results are extracted to complete the processing quality evaluation;Finally,a certain type of blade is taken as the experimental object.The test results are compared with the three coordinate measurement results.The results show that the profile deviation of the three target sections is within 0.02 mm,the maximum deviation of feature parameters is 0.017 mm,and the maximum relative deviation of feature parameters is 2.22%.It shows that the proposed method can calculate the profile deviation and characteristic parameter deviation of the blade to be tested,so as to judge whether the blade to be tested is qualified,which verifies the feasibility of the method.(5)The method of rapid reconstruction of blade profile was studied.Taking the repaired aeroengine blade as an example,by setting the sampling frequency of the line structured light sensor and the movement speed of the Z axis,the rapid scanning of the blade surface is realized.The 3D shape of the blade was reconstructed by using reverse engineering software to process the point cloud data such as noise reduction and curvature sampling.The reconstruction scheme can be widely used in the fields of blade machining,repair,inspection and so on. |
