基于参数优化VMD的齿轮箱故障特征提取方法

Fault Feature Extraction Method of Gearbox based on Parameter Optimization VMD

为解决齿轮箱故障振动信号信噪比低、故障特征提取难的问题,提出了基于参数优化变分模态分解(VMD)的齿轮箱故障特征提取方法。首先,以分解结果的局部极小包络熵最小为目标,利用果蝇算法搜寻VMD分解参数K和α的最优组合;将原始信号分解成若干IMF分量,从中选择包络熵较小的分量进行信号重构,并对重构信号进行包络解调运算,从重构信号的包络谱中提取故障频率特征。结果表明,利用此方法对实测信号进行处理,成功降噪、提取齿轮箱故障特征,并且比利用经验模态分解方法降噪效果更好,提取的故障特征更加明显。

In order to solve the problem that the signal-to-noise ratio of the gearbox fault signal is low and fault feature extraction is difficult,a method for extracting gearbox fault feature based on parameters optimized variational mode decomposition is proposed.Firstly,the drosophila optimization algorithm is used to search for the most optimal combination of the variational mode decomposition's K andα,aiming at the minimum local en⁃tropy of the decomposition result.The original signal is decomposed into several IMF components,from which the component with the smaller envelope entropy is selected for signal reconstruction,and the reconstructed sig⁃nal is demodulated to extract the fault frequency feature from the envelope spectrum of the reconstructed signal.The results show that this method can reduce the noise and extract the fault features of gearbox successfully,and the effect of noise reduction is better than the empirical mode decomposition method,and the extracted fault fea⁃tures are more obvious.