To solve the issue of bearing signal fault features being easily drowned out by noise,a parameter optimization variational mode decomposition(VMD)method combined with improved wavelet packet threshold denoising is proposed.Firstly,by using VMD combined with Improved Particle Swarm Optimization(IPSO),the noisy signals are decomposed into several Intrinsic Mode functions(IMF).Based on the criterion of maximum correlation coefficient-correlation kurtosis,the IMFs are divided into high-value Intrinsic Mode functions(HIMF)and low-value Intrinsic Mode functions(LIMF).The Improved wavelet packet threshold denoising is applied to the LIMF.Finally,the envelope of the reconstructed signal is demodulated,and the bearing fault characteristic frequency is extracted.And the fault diagnosis is completed.Experimental results show that the proposed method can not only overcome the phenomenon of"overkill",but also obtain the denoised signal with higher signal-to-noise ratio.
关键词
振动与波/变分模态分解/小波包阈值去噪/相关峭度/相关系数/轴承
Key words
vibration and wave/variational mode decomposition/wavelet packet threshold denoising/correlation kurtosis/correlation coefficient/bearing
维普
万方数据