基于经验模式分解的包络解调技术及其应用

提出了一种新的调制信号解调分析技术,来抑制传统的包络解调方法中经常出现的无意义的频率成分.首先,对复杂的振动信号进行经验模式分解,得到若干个基本模式分量,再对包含调制信号的基本模式分量进行包络分析以提取故障信息.该方法利用经验模式分解来实现故障信息的有效分离,从而提高了诊断信号的信噪比.利用该方法对某齿轮箱轴承座振动信号进行经验模式分解,进而解调出高速轴转频这一调制频率,准确地诊断出该故障是由齿轮轴不对中所引起的,通过针对性的维修后,消除了故障,从而验证了该方法的有效性.

Envelope Demodulation Method Based on Empirical Mode Decomposition with Application

A new envelope demodulation method based on empirical mode decomposition (EMD) is proposed to make up the limitation that meaningless frequencies often arise in the envelope spectrum of conventional envelope demodulation method. The vibration signal is decomposed into several intrinsic mode functions (IMFs) by EMD, and the fault information is extracted from the IMFs containing modulated signals, which then is effectively separated by EMD, thereby the signal to noise ratio (SNR) of diagnosis signal is improved obviously. With the aid of the proposed method, vibration signal of one gear-box bearing bridge is decomposed by EMD, and the modulated frequency corresponding to the rotating frequency of high speed axis is demodulated to affirm the fault due to the misalignment of two gear axes. A proper maintenance removes the fault thoroughly, which verifies the validity of this method.