基于参数优化变分模态分解和多尺度熵偏均值的行星变速箱故障特征提取

针对某型装甲车辆行星变速箱行星轮故障特征难以提取的问题，提出了结合参数优化变分模态分解(VMD)和多尺度熵偏均值的故障特征提取方法。为克服VMD算法参数选取依赖经验的弊端，采用粒子群优化算法对VMD参数进行优化。使用参数优化后的VMD算法对信号进行分解，并依据互信息选取有效分量对信号进行重构。多尺度熵能反映信号在多尺度下的复杂度，偏均值可以反映多尺度熵的均值和变化趋势。采用基于多尺度熵的综合指标多尺度偏均值，以全面反映振动信号在多尺度下的特性，用于衡量行星变速箱运行状态，从而进行故障特征提取。行星变速箱实验数据处理结果表明，新方法可以更加有效的提取行星变速箱故障。

Planetary Gearbox Fault Feature Extraction Based on Parameter Optimized Variational Mode Decomposition and Partial Mean ofMulti-scale Entropy

A fault feature extraction method based on parameter optimized variational mode decomposition（VMD）and partial mean of multi-scale entropy is proposed for the extraction of planetary gearbox fault features. The particle swarm optimization（PSO） is used to optimize VMD parameters to overcome the drawback of subjectively selecting the parameters, and the signal is processed by parameter-optimized VMD. The effective components are selected by mutual information to reconstruct signal. Multi-scale entropy can be used to analyze the complexity of signal under different scale factors, and partial mean can be used to reflect the mean value and variation trend of data. Based on multi-scale entropy, the partial mean of multi-scale entropy is used to measure the operating states of planetary gearbox, thus extracting the fault features. The processed results of planetary gearbox experimental data show that the proposed method can be more effective for planetary gearbox fault extraction.