煤矿机械齿轮箱故障诊断方法

Fault diagnosis method of coal mine machinery gearbox

  • 摘要: 针对煤矿机械齿轮箱振动信号中含有大量噪声干扰成分导致齿轮箱故障特征提取难的问题,提出了一种基于粒子群优化变分模态分解(PSO-VMD)与最小熵反褶积(MED)的煤矿机械齿轮箱故障诊断方法。该方法首先利用PSO算法对VMD中直接影响分解效果的惩罚系数与分量个数进行优化搜索,得到最大化分解性能的最优参数组合,并利用优化后的VMD方法对齿轮箱振动信号进行分解,得到一系列本征模态函数(IMF)分量;然后,利用MED方法对与原信号相关度最大的IMF分量进行降噪处理,凸显故障冲击特征;最后,对降噪后的IMF分量进行Hilbert包络解调,从而提取故障特征。实验结果表明,该方法能够准确提取故障特征,实现齿轮箱故障诊断。

     

    Abstract: In order to solve problem that vibration signal of coal mine machinery gearbox contains a large number of noise interference components, which makes it difficult to extract fault characteristics of the gearbox, a fault diagnosis method of coal mine machinery gearbox based on particle swarm optimization variational mode decomposition(PSO-VMD) and minimum entropy deconvolution(MED) is proposed. Firstly, PSO algorithm is used to optimize search for punish coefficient and the number of components directly affecting decomposition effect in VMD, so as to obtain the optimal parameter combination to maximize decomposition performance. The optimized VMD method is applied to decompose gearbox vibration signal to obtain a series of intrinsic mode function(IMF) components. Then, IMF component with the highest correlation with original signal is denoised by MED method to highlight fault impact characteristics. Finally, Hilbert envelope demodulation is performed for IMF components after noise reduction, so as to extract fault characteristics. The experimental results show that the method can accurately extract fault characteristics and realize gearbox fault diagnosis.

     

/

返回文章
返回