自适应MED降噪和EMD分解在注塑机轴承故障诊断中的应用  

Application of Adaptive MED Noise Reduction and EMD Decomposition in Fault Diagnosis of Mechanical Bearings

作　　者：李水利[1] 周建民 孙定邦 LI Shui-li;ZHOU Jian-min;SUN Ding-bang(Shanxi Institute of Mechanical&Electrical Engineering,Changzhi 046011,China;Huaihai Industrial Group Co.,Ltd.,Changzhi 046012,China;Taiyuan University of Technology,Taiyuan 030000,China)

机构地区：[1]山西机电职业技术学院,山西长治046011 [2]淮海工业集团有限公司,山西长治046012 [3]太原理工大学,山西太原030000

出　　处：《塑料科技》2020年第6期124-128,共5页Plastics Science and Technology

摘　　要：针对注塑机轴承故障信号含有强背景噪声且难以诊断的问题,提出使用最小反褶积(MED)降噪算法结合经验模态分解(EMD)对注塑机轴承的故障振动信号进行分析,通过所得频谱确定注塑机轴承故障位置。结果表明:该方法可以对含有较大背景噪声的仿真信号及实例轴承信号去噪,经EMD分解后可得到高信噪比故障信号,通过频谱分析技术可以很好地诊断输入信号是否存在故障。In view of the problem that the injection machine bearing fault signal contains strong background noise and is difficult to diagnose,it is proposed to use the minimum deconvolution(MED)noise reduction algorithm combined with empirical mode decomposition(EMD)to analyze the vibration signal of the injection molding machine bearing fault,and determine the fault location of the injection molding machine bearing through the obtained frequency spectrum.The results show that this method can denoise the simulated signal with large background noise and the bearing signal of the example.After EMD decomposition,a high signal-to-noise ratio fault signal can be obtained.The spectrum analysis technology can be used to diagnose whether the input signal is faulty.

关 键 词：注塑机 故障诊断 最小反褶积 经验模态分解 频谱分析 

分 类 号：TP3063[自动化与计算机技术—计算机系统结构]