基于EMD的FCMM降噪和故障诊断

摘要

为了解决现场三坐标测量机(FCMM)在加工车间现场环境下由于振动、环境因素等干扰造成的输出带有较强背景噪声而可能导致滚动轴承发生故障的问题，提出一种基于经验模态分解(EMD)的 FCMM降噪和故障诊断方法，对 FCMM输出进行 EMD分解得出本征模态函数(IMF)分量，采用互相关系数准则法对 IMF 分量降噪重构，同时进行边际谱分析，从中提取故障特征频率。实验根据 FCMM速度参数测量数据，结合 Matlab 仿真方法，得到 FCMM输出速度数据的 IMF分量与原始速度数据之间的互相关系数表以及 IMF分量边际谱图。实验结果验证了 EMD方法滤波效果明显，故障分析手段可靠。%In the field environment such as processing workshop,vibration and other field complex environmental interfer-ences make field coordinate measuring machine (FCMM)output with a strong background noise,and may result in FCMM rolling bearing failure.FCMM de-noising and diagnosis method based on empirical mode decomposition (EMD)is proposed. Intrinsic mode function (IMF)component is derived from FCMM output by using EMD,then IMF is reconstructed by com-paring cross-correlation,at the same time marginal spectrum analysis is used to extract fault characteristic frequency.Ac-cording to FCMM speed data,the table of cross-correlation between original speed data and IMF component and the map of marginal spectrum are obtainedt by Matlab simulation.The experimental results verify that EMD has obviously filtering effect and reliable failure analysis.