对飞机发动机转子系统早期故障特点进行分析的基础上,提出了利用虚拟仪器和M atlab小波工具箱分析软件对其早期故障进行检测和特征提取的方法。文中对早期故障特征量的选取、有用信号与噪声信号的分离方法、突变信号与奇异信号的特征提取等进行了分析和研究,并以转子早期碰摩和早期不平衡为例进行了实验研究。结果表明,Labview和M atlab小波分析软件相结合,能够快速有效地提取发动机转子系统的早期故障特征,为发动机转子系统早期故障的快速识别提供了一种新途径。

A method to detect early fault and extract its features of aero-engine rotor systems with VI(Virtual Instruments) and MT(Matlab Tools) is proposed.It is based on the analysis of early fault of the aero-engine rotor systems.The method is described in detail.The early fault characteristic signal from detected signals and analysis the selection of early fault eigenvalue is extracted,in which useful signal is separated from dominant noise signal.The experiments on the early rub-impact and early unbalance of the aero-engine rotor system is performed and the effectiveness of our method was demonstrated in the following respects: it shows that not only complex calculation can be avoided with VI and MT but also early fault features can be quickly extracted.The early fault feature signal can be obtained by means of de-noising,decomposition and reconstruction.Early fault features of aero-engine rotor system can be fast extracted from the complex signals by energy distribution of feature frequency.Early fault can be separated and amplified by Wavelet Analysis.

V263.6

国家自然科学基金(60472116);航空科学基金资助(03I53068)