基于噪声信号和改进VMD的滚动轴承故障诊断

受运行环境及传递路径影响,滚动轴承声音信号中包含有强背景噪声和较大的非周期性瞬态冲击成分,导致轴承故障特征提取困难。文中提出一种基于自适应变分模态分解(AVMD)的滚动轴承噪声信号故障诊断方法。该方法首先根据不同的信号自适应地确定模式数和惩罚因子,利用优化参数的VMD对原始信号进行分解,得到多个本征模式分量;然后计算各模式分量时域、包络谱和时-频加权峭度,根据时-频加权峭度最大化准则选择最佳IMF;最后采用共振解调技术求出最佳IMF包络谱。对轴承故障信号研究表明,所提方法可解决传统VMD算法分解精度受参数影响较大,导致信号出现过分解或欠分解的问题。另外与传统方法相比,该方法可以在强背景噪声和非周期性瞬态冲击下有效识别轴承故障。

Fault Diagnosis of Rolling Bearings Based on Noise Signal and Improved VMD

Affected by the operating environment and transmission path,the sound signals of rolling bearings usually contain strong background noise and large non-periodic transient impact components,which makes it difficult to extract the fault features of the bearings.In this paper,an adaptive variational mode decomposition(AVMD)method for fault diagnosis of rolling bearings based on noise signals is proposed.First of all,the number of modes and penalty factors are adaptively determined according to different signals,and the optimized parameter VMD is used to decompose the original signals to obtain multiple intrinsic mode components.Then,the time domain,envelope spectrum and time-frequency weighted kurtosis of each modal component are calculated.The best IMF is selected according to the time-frequency weighted kurtosis maximization criterion.Finally,the resonance demodulation technique is used to find the envelope spectrum of the best IMF.The research shows that the proposed method overcomes the problem that the decomposition accuracy of traditional VMD algorithm is greatly affected by the parameters,resulting in over-decomposition or under-decomposition of signals.In addition,compared with traditional methods,this method can effectively identify bearing faults under strong background noise and non-periodic transient impact.