Data Processing Of Aviation Blade Inspection And Testing Software Development

The aero blade is an important power component of an aero engine.It has been used in high temperature(high temperature end > 2000K),high pressure,alternating load,high-speed rotation,high-frequency vibration and high-temperature gas corrosion under long-term service.Twisted,thin-walled,hollow,film cooling holes,integral blade disk are too difficult to ensure the accuracy of blade shape and position during casting and milling.How to achieve rapid and high-precision inspection of aviation blades has always been a hot spot in aviation manufacturing.This article focuses on the research of aviation blade inspection methods,technical specifications,data processing algorithms,and inspection software development.The main contents are as follows:(1)Study the basic classification,structure,key surface profile parameters and profile modeling methods of aero blades,focusing on aero blade inspection technical specifications,key steps and inspection parameters,applying aero blade design theory and combining aero blade inspection practices,the evaluation methods of blade shape,size and position error are proposed to solve the problems of inconsistent evaluation standards and irregular process of the aviation blade inspection.(2)Study the principle of aero blade measuring point data segmentation,cluttered point cloud sorting and profile feature parameter extraction,and use K-means++ clustering to propose a multi-blade and multi-section point cloud segmentation algorithm to overcome the point cloud under segmentation and over segmentation,using outlier detection,concave hull construction and depth-first search,propose an algorithm to identify and delete noise inside and outside the contour and sort the point cloud,realized a fast curve reconstruction of disordered point cloud,and propose profile segmentation and the feature parameter extraction algorithm,and realize accurate profile segmentation and rapid feature parameter extraction.(3)Participate in the development of i Point3 D Blade,a data processing software for aviation blade measuring points,build major modules such as displaying,point cloud processing,parameter calculation and report output,design point cloud data format IPD,and integrate a variety of point cloud matching algorithms such as ICP,TDM and ADF to achieve fast and high-precision matching of point clouds of different types,data scales and initial pose,and develop automated parallel batch operations function,and greatly improve the calculation efficiency of large-scale measuring point data processing.(4)Using 6 samples such as cast turbine blades,nozzle guide vanes and compressor blades,compared with traditional manual multiple commercial software interactive operations,carried out blade data processing experiments and quantitatively studied evaluation indicators such as the profile error,profile deviation,torsion angle,overall average error and average inspection time.The results of the experiment show that the i Point3 D Blade software achieves a better overall average error and a single-section blade inspection time of less than 30 s and saves 80% of the inspection time.