粒子滤波在轴承故障振动信号降噪中的应用

Application of Particle Filter in Rolling Bearing Fault Vibration Signal Denoising

针对滚动轴承振动信号容易受到较为复杂的随机噪声的污染,提出了基于Rao-blackwellised粒子滤波的振动信号降噪方法。建立了不含噪的振动信号的时变自回归模型,进而转化成对应的状态空间模型,把降噪问题转化成在状态空间模型下的滤波问题,并用仿真信号进行了试验研究,结果表明,该方法具有较好的降噪效果。

Rolling bearing vibration signal is vulnerable to be submerged in complex random noise.A vibration signal denoising method is presented based on Rao-blackwellised particle filtering.TVAR model of the clean vibration signal is established and state the vibration signal in a state-space form.Then the assignment of denoising is treated as a filter problem.Synthetic data and real vibration signal tests are carried out to investigate the effectiveness of the suggested algorithm.