基于EMD-ICA去噪的滚动轴承故障诊断方法  

Fault diagnosis of rolling bearing based on EMD-ICA de-noising

作　　者：蔡剑华[1] 胡惟文[1] 王先春[1] 

机构地区：[1]湖南文理学院信息研究所,湖南常德415000

出　　处：《机械设计》2015年第1期17-23,共7页Journal of Machine Design

基　　金：国家自然科学基金资助项目(41304098);湖南省自然科学基金资助项目(12JJ4034);湖南省教育厅青年基金资助项目(13B076);湖南省重点建设学科--光学基金资助项目(11GX020);湖南省重点实验室"光电信息集成与光学制造技术"资助项目

摘　　要：针对轴承故障信号易受环境噪声影响、信噪分离难的问题,提出了一种基于经验模态分解和独立成分分析相结合去噪的滚动轴承故障诊断方法。给出了该方法在故障诊断信号去噪领域的应用原理、方法步骤和评价指标;并通过仿真信号和实际轴承的滚动体故障、内圈故障和外圈故障信号进行了分析和故障诊断,验证了该方法在轴承故障信噪分离中的有效性。结果表明,采用文中提出的方法消噪后提取故障信号特征频率,压制了噪声干扰,能明显区别出轴承的状态及其故障的类型,有效提高了轴承故障诊断的准确性。A method of fault diagnosis was proposed which combined empirical mode decomposition with independent component analysis based on the facts that the fault signal of rolling bearings was affected easily by environment noise and difficult to be separated from noise.The principle and steps of the method were given and the de-noising effect was evaluated by some parameters.The simulated signal and actual fault signals of the bearing ball,the inner and outer race were analyzed and diagnosed.The results show that the proposed method can suppress the noise interference greatly.Fault characteristics extracted from the de-noised signal can distinguish the state of bearing and type of faults obviously.The accuracy of fault diagnosis of the bearing is improved effectively.

关 键 词：经验模态分解 独立分量分析 滚动轴承 故障诊断 去噪 

分 类 号：TH17[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]