摘要
针对现有各种降噪方法处理非平稳机械振动信号存在的缺点,提出一种基于辅助白噪声经验模式分解技术来自适应实现旋转机械非平稳振动信号降噪。该方法是一种集成的经验模式分解(Ensemble Empirical mode decomposition,EEMD)降噪算法,利用正态分布白噪声在经验模式分解中具有的二进尺度分解特性,可以有效抑制常规经验模式分解降噪算法处理非平稳振动信号时产生的模式混叠现象。通过仿真计算和转子启动过程试验振动信号对新降噪方法、经验模式分解降噪方法及小波降噪方法的性能进行了比较测试,结果表明,在非平稳机械振动信号降噪方面,新降噪方法具有更高的信噪比,不仅能够消除高斯噪声,而且能够有效降低脉冲干扰,提取出反映信号实际物理意义的振动固有模式。
Aiming at defects of different de-noising methods in processing non-stationary vibration signals,a method based on Gauss white noise assisted empirical mode decomposition (EMD) technique,which adaptively eliminates noise involved in non-stationary vibration signals of large rotating machines was presented. This method was an ensembled algorithm of EMD de-noising method,it applied dyadic scale decomposition characteristics of the normal distribution white noise in EMD,it could effectively suppress mode mixing that often occurred in analyzing non-stationary vibration signals with EMD de-noising method and could obtain a higher signal-to-noise ratio(SNR). Numerical simulation signals and experimental signals of rotor running state were used to test and compare the performances of the proposed method,the EMD-based de-noising method and the wavelet de-noising method. The results showed that the EEMD-based noise cancellation method presented here had more effective de-noising performance,not only eliminated random noise,but suppressed impulse noise and extracted vibration intrinsic modes that reflect real physical meaning of the signal.
出处
《振动与冲击》
EI
CSCD
北大核心
2009年第9期33-38,共6页
Journal of Vibration and Shock
基金
国家自然科学基金(50675194)
国家863(2008AA04Z410)资助项目
关键词
降噪
旋转机械
启动过程
振动信号
集成经验模式分解
de-noising
rotating machine
running state
vibration signal
ensemble empirical mode decomposition(EEMD)
