为实现低信噪比下的微弱信号检测,提出一种基于局域波和混沌的微弱信号检测方法.将微弱的故障信号分解为有限的并且具有不同基本模式的分量,每个分量为单一成分信号,实现了信噪分离.将局域波分量输入所设计的混沌振子,混沌振子系统行为由混沌状态变为大周期运动状态,表明检测信号中含有特征成分,实现了利用混沌振子对低信噪比微弱信号的检测识别.对转子系统早期碰摩故障信号检测结果说明了该方法的有效性.
A weak signal detection method combining local wave method and chaos was developed to realize weak signal detection in low SNR. The weak fault signal can be decomposed into finite local wave components with different simple intrinsic modes, so the signal was separated from noise. The components were input into chaos oscillator, which was transformed from chaotic state to large-scaled periodic state, so that the weak signal can be identified in low SNR through distinguishing the characteristic frequency of signal. Test on early rub-lmpact fault diagnosis for rotor system shows the validity of the presented method.