基于EMD-PSO-MKD的滚动轴承故障特征提取  

Rolling Bearing Fault Feature Extraction Based on EMD-PSO-MKD

作　　者：段宜征 DUAN Yizheng(Shenyang University of Chemical Technology,Shenyang 110142,China)

机构地区：[1]沈阳化工大学机械与动力工程学院,沈阳110142

出　　处：《机械工程师》2025年第4期53-56,共4页Mechanical Engineer

摘　　要：针对滚动轴承早期故障特征提取不充分的问题。提出了一种采用经验模态分解(EMD)与粒子群算法(PSO)优化最大时域峭度解卷积(MKD)相结合的滚动轴承故障特征提取新方法。首先,采用EMD将原始振动信号分解成若干固有模态分量(IMF),然后选择峭度和互相关系数准则筛选出高频振荡部分用以重构信号;其次,利用时域峭度作为粒子群算法的适应度函数以选择MKD最优参数组合;最后,将优化后的MKD应用在滚动轴承,完成整个故障特征提取工作。应用凯斯西储大学数据集分析表明,该方法不仅可以有效地提取故障特征,而且诊断成功率高达100%。In view of insufficient extraction of early fault features of rolling bearings,this paper proposes a new method for extracting rolling bearing fault features by combining empirical modal decomposition(EMD)and particle swarm optimization(PSO)optimization maximum time domain steepness deconvolution(MKD).Firstly,EMD is used to decompose the original vibration signal into several intrinsic mode components(IMF),and then select the steepness and correlation number criteria to screen out the high-frequency oscillation part to reconstruct the signal.Secondly,the time-domain steepness is used as the fitness function of the particle swarm algorithm to select the optimal parameter combination of MKD.Finally,the optimized MKD is applied to the rolling bearing to complete the entire fault feature extraction work.The application of Case Western Reserve University dataset analysis shows that the method not only effectively extracts fault features,but also has a diagnostic success rate of up to 100%.

关 键 词：滚动轴承 经验模态分解 PSO MKD 故障诊断 

分 类 号：TH165.3[机械工程—机械制造及自动化] TN911.7[电子电信—通信与信息系统]