Research On Defects Identification Of Engine Blade Based On Mixed Reality Platform

As the "heart" of the aircraft,the health and safety of the aero-engine must be guaranteed100%.Many factors are causing the engine failure,among which the failure caused by the damage of the engine blade accounts for a large proportion of all engine failures.Therefore,the defects inspection of the engine blade is directly related to the safe operation of the aircraft.As an important part of aero-engine,the manufacturing process of engine blades is complex,and there are many kinds of blades.How to efficiently and quickly check the blade defects has been plagued by front-line maintenance personnel and Repair personnel.Based on this research background,to reduce the difficulty of blade defects investigation,adhering to the concept of intelligent maintenance,this paper first introduces the composition of engine blade system in detail,especially the material and function of low-pressure compressor blade,high-pressure compressor blade,high-pressure turbine blade and low-pressure turbine blade which directly affect the engine performance The common faults of each part of the blade and the possible factors causing the faults are described,and the recognition characteristics of common typical faults are summarized,to facilitate the fault recognition of machine vision.The inspection standard of engine blade is mainly composed of two parts: the standard of borescope inspection in Aircraft Maintenance Manual(AMM)and the standard of visual inspection in Engine Shop Manual(ESM).According to the manual,the defects inspection standards of each part of the blade were classified and studied,the leaf partition mode of each part of the blade was introduced,the manual standards were sorted out,and the blade defects inspection standard system was constructed.This paper proposes that blade defects recognition based on mixed reality platform is an ideal framework for the application.It mainly researches and realizes the positioning module of typical blade defects,and uses the Canny edge detection algorithm and Hough transforms two image processing methods.At the same time,it also realizes the augmented reality recognition and pre-scan fault labeling of the blade and compiles the logic control code of the blade defects inspection standard system.Based on the unity3 D mixed reality platform,it integrates the workshop standard of blade defects inspection to identify the typical defects of the engine blade,strictly abides by the manual standard,and simplifies the manual query process,which has a certain engineering application value.