针对轴承振动信号夹杂的噪声极大地影响有用信息的提取,提出了基于遗传算法的变分模态分解(Variational mode decomposition,VMD)与小波阈值去噪方法.该方法首先利用遗传算法选择合适的VMD参数,然后用VMD方法对含噪声的信号进行自适应分解,最后对分解的模态分别进行小波阈值处理后重构信号,得到去噪后的信号.对实际轴承信号的分析结果表明,该方法与常用的去噪方法相比,能够得到更高的信噪比和更低的均方差.%Aiming at the extraction of useful fault feature information from bearing vibration signal affected by the noise,the variational mode decomposition (VMD) and wavelet threshold denoising method based on genetic algorithm is proposed.The method firstly utilizes genetic algorithm selecting appropriate parameters of the VMD,then the noise signal is decomposed adaptively by VMD method,finally processing the modes of decomposition respectively by wavelet threshold method,restructuring the signal to get denoised signal.Experimental results on actual bearing signals show that the proposed method can obtain higher signal-to-noise ratio and lower mean square deviation compared with several common denoising methods.
展开▼
