基于涡流检测信号的航空发动机叶片缺陷分类与评估方法  

Study on Classifying and Evaluating Defects of the Aviation Engine Blade Based on Eddy Current Detection Signals

作　　者：于霞[1] 张卫民[1] 邱忠超[1] 秦峰[1] 

机构地区：[1]北京理工大学机械与车辆学院,北京100081

出　　处：《测试技术学报》2016年第2期99-105,共7页Journal of Test and Measurement Technology

基　　金：总装预研基金资助项目(9140A17080610BQ×××)

摘　　要：航空发动机涡轮叶片的缺陷检测对于保障飞机安全运行至关重要.由于叶片属于非规则小曲率零件,难以保证严格的提离距离和检测法向方向,由此产生了不可忽视的干扰和噪声,加之缺陷变化信息微弱,给检测带来了实质性困难.本文设计研制了一种尺寸小、灵敏度高的差激励涡流检测探头,可以安装在数控多自由度扫查台上,对叶片曲面零件表面缺陷进行快速扫查检测;利用总体平均模态经验模态分解技术(EEMD)和小波变换相结合的方法,来有效抑制强背景噪声,提取信号特征,并结合支持向量机(SVM)方法实现裂纹缺陷的分类.It is very important to detect blade defects of aero engine turbine for protecting the safety of aircraft operation.Because the blade belongs to irregular curvature parts,It is difficult to ensure strict lift off distance and normal detection,which results in considerable interference and noise since the blade belongs to irregular curvature parts,otherwise weak defect information brings substantive difficulties to detection.A differential incentive eddy current detection probe of small size and high sensitivity is designed and developed,which can be installed on the CNC multi-freedom scanning table and detect defects on blade surface parts rapidly;The combination of overall average modal and empirical mode decomposition technique（EEMD）as well as wavelet transform were used to suppress the strong background noise effectively and extract signal feature,then combined with support vector machine（SVM）method to achieve defect classification.

关 键 词：发动机叶片 差激励 涡流传感器 裂纹检测 EEMD SVM 

分 类 号：TG115.28[金属学及工艺—物理冶金]