高精度频率分辨在振动信号辨识中的应用

(1. 北京航天动力研究所,北京 100076;2. 北京宇航系统工程研究所,北京 100076)

抗混叠低通滤波;频率分辨率;信号抽取;短时傅里叶变换

Application of high-precision frequency resolution in vibration signal identification
ZHANG Zhaolei1, SUN Xiaowei1, XIA Dexin1, LI Fang2

(1. Beijing Aerospace Propulsion Institute, Beijing 100076, China;2. Beijing Institute of Aerospace Systems Engineering, Beijing 100076, China)

anti-aliasing low-pass filtering; frequency resolution; signal extraction; short-time Fourier transform

备注

发动机振动信号处理通常采用DASP软件进行。由于软件功能限制,DASP处理采样频率为50 kHz发动机振动信号时,频谱的频率分辨率最高精度仅为12.2 Hz,导致对微弱及故障信号的辨识较为困难。本文采用抗混叠低通滤波、信号抽取及短时傅里叶变换相结合的信号分析方法,有效提高了频率分辨率的精度,最高精度达到1.2 Hz,达到了理想的频率辨识效果。

DASP software is usually adopted to conduct vibration signal processing of rocket engine. Due to the limitation of the software function, the highest accuracy of spectrum frequency resolution is merely 12.2 Hz when the sampling frequency processed by DASP is 50 kHz engine vibration signal, which may cause a difficulty of distinguishing the weak and fault signals. A new signal analysis method that combines anti-aliasing low-passes filtering, signal extraction and short-time Fourier transform is adopted in this paper, which can improve the accuracy of frequency resolution effectively. The highest accuracy of the method can reach up to 1.2 Hz, which achieves the ideal effect of frequency identification.