摘要
针对RV减速器振动信号中含有大量噪声干扰成分导致减速器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)与最大相关峭度解卷积(PSO-MCKD)的RV减速器故障诊断方法。采用粒子群算法对变分模态分解中的分解层数c、惩罚因子α、最大相关峭度解卷积的滤波器长度L及位移参数M进行参数寻优,以多尺度排列熵(PME)为适应性函数,得到最优分解组合。对采集到的振动信号进行变分模态分解,通过分解得到系列本征模态函数(IMF)分量;以峭度值为筛选准则,计算出与原信号相关度最大的IMF分量,利用得到的最优[L,M]值对原始信号进行最大相关峭度解卷积,凸显故障冲击特征;对降噪后的IMF分量进行希尔伯特包络解调,从而提取故障特征。同时进行对比试验,验证了该方法的优越性。试验结果表明:该方法能够准确提取故障特征,减小噪声的影响,实现RV减速器的故障诊断。
Since it is difficult to extract fault features due to a large amount of noise interference components in the vibration signal of RV reducers,in this article a method of fault diagnosis is proposed for RV reducers,which is based on particle swarm optimization variational mode decomposition(PSO-VMD)and maximum correlation kurtosis deconvolution(PSO-MCKD).The particle swarm optimization algorithm is used for parameter optimization of the decomposition level c and the penalty factorαin variational mode decomposition,as well as the filter length L and the displacement coefficient M of the maximum correlation kur-tosis deconvolution;with the multi-scale permutation entropy(PME)as the adaptive function,the optimal decomposition combina-tion is obtained.Furthermore,the collected vibration signals are subject to variational mode decomposition,in order to work out a series of intrinsic mode function(IMF)components through decomposition.With the kurtosis value as the screening criterion,ef-forts are made to calculate the IMF components with the highest correlation with the original signal;then the obtained optimal[L,M]value is used to perform maximum correlation kurtosis deconvolution on the original signal,thus highlighting the fault impact charac-teristics.Hilbert envelope demodulation is performed on the denoised IMF components to extract fault features.At the same time,comparative experiments are conducted to verify the superiority of this method.The experimental results show that this method can accurately extract fault features,suppress the influence of noise,and achieve fault diagnosis of RV reducers.
作者
罗捷
蔺梦雄
LUO Jie;LIN Mengxiong(College of Electrical Engineering,Guangxi Technological College of Machinery and Electricity,Nanning 530007;College of Urban Transit and Logistics,Beijing Union University,Beijing 100101)
出处
《机械设计》
北大核心
2025年第9期170-176,共7页
Journal of Machine Design
基金
广西教育科学规划研究课题(2022ZIY2251)。
关键词
RV减速器
故障诊断
粒子群优化算法
改进变分模态分解
最大相关峭度解卷积
多尺度排列熵
RV reducer
fault diagnosis
particle swarm optimization algorithm
improved variational mode decomposition
maximum correlation kurtosis deconvolution
multi-scale permutation entropy
