Research On Three-dimensional Blade Measurement Method Based On Robot And Line Laser Scanning

Aero-engine is the core of China's aviation industry development,and it is the first of the "Top 100 Projects" of China's 13 th Five-Year Plan.It has great significance in the fields of national defense construction and military science.The blade is a typical free-form surface part.As the largest number of parts in an aero engine,it is subjected to high temperature and high pressure during the operation of the aero engine.It also needs to run stably and continuously at high speed.The manufacturing level of the blade directly determines the aerodynamic performance of the aero engine,which puts extremely strict requirements on the processing procedures of the blade.Therefore,the efficient and rapid detection of the processing quality of the blade has become an urgent problem to be solved in the field of aero engine development.A fast and efficient three-dimensional blade measurement and profile detection method is of great significance.In response to the above problems,this subject proposes a three-dimensional measurement method of aero-engine blades based on robot and line laser scanning,which realizes fast and efficient measurement of the blades,and an algorithm for extracting the characteristic parameters of the blade profile(including front and rear edges,chord length,and mid-arc Line,maximum thickness,etc.)to conduct in-depth research to solve the problem of blade machining quality evaluation,the main research contents are as follows:This subject proposes a system measurement model in which a six-degree-offreedom robot drives a line laser sensor to scan the blade.The D-H method was used to establish the robot kinematics model so that the coordinate conversion matrix of the robot end flange coordinate system to the base coordinate system can be calculated.The measurement principle of the line laser sensor is analyzed.The adapter that fixes the line laser sensor at the end of the robot is designed and processed.Eye-in-hand calibration method is researched to calculate the coordinate conversion matrix from the sensor coordinate system to the robot end flange coordinate system.Then the two-dimensional contour data measured by the line laser sensor can be converted into the three-dimensional point cloud data of the blade in the base coordinate system.For the measured point cloud data of the blade,the point cloud processing algorithm based on the Point Cloud Library is used to realize the three-dimensional reconstruction of the blade.The noise data in the point cloud is eliminated by the noise filtering method based on statistical analysis technology.The down-sampling method of the prime grid downsamples the point cloud data to increase the speed of the algorithm and complete the preprocessing of the point cloud data.This subject studies the nearest neighbor search method of point cloud data based on KD-tree and the calculation method of normal vector of point cloud data,and uses greedy projection triangulated surface reconstruction algorithm and Poisson surface reconstruction algorithm to reconstruct the blade three-dimensionally.Principal component analysis is used to intercept the blade point cloud data from the measured point cloud.The polyline dichotomy is used to sort the point cloud of the profile,and the curve inspection method is used to denoise the point cloud of the profile.The minimum area straight line is used to approximate Method to extract approximate chords and partition the blade profile point cloud.The front and rear edge feature is extractioned by the algorithm based on the least square ellipse fitting method.The front and back edge points can be accurately extracted.The projection method is used to extract chord length parameters.The analytical method based on equal radius idea is used to extract the mid-arc parameters the maximum thickness.Finally,the blade measurement scheme proposed in this paper is designed and implemented.The measurement system is built and the trajectory control of the robot is realized using the ROS platform.The synchronous collection and processing of line laser measurement profile data and robot joint angle data are realized under Linux system.Three-dimensional reconstruction of the measured point cloud data through the Point Cloud Library and comparison with commercial software.Intercepting the model point clouds at different heights from the measured point cloud data,and the feature parameters of the profile are extracted using the algorithm proposed in this paper.The result is compared with traditional algorithms to verify the effectiveness of the extraction algorithm proposed in this topic.