拟合故障振动信号模型实现滚动轴承故障诊断  

Fault Diagnosis of Rolling Bearing by Vibration Signal Model Fitting

作　　者：郭艳平[1] 解武波 龙涛元[1] 

机构地区：[1]中山火炬职业技术学院,广东中山528436 [2]中山天誉真空科技有限公司,广东中山528436

出　　处：《机械设计与制造》2017年第11期205-208,共4页Machinery Design & Manufacture

基　　金：国家自然科学基金(61273168);中山市科技计划项目(2013A3FC0309)

摘　　要：针对传统故障诊断流程的缺点,提出通过拟合故障振动信号模型实现滚动轴承故障诊断,并在风力发电机组齿轮箱故障诊断中验证了有效性和实用性。首先根据滚动轴承发生故障时振动信号的特点,提出故障振动信号模型,然后通过遗传算法对该模型做数据拟合,拟合数据来自EMD(Empirical Mode Decomposition)方法对原始振动信号分解所得IMF(Intrinsic Mode Function)分量,最后将拟合结果和轴承各部件的故障特征频率作对比,可知损伤点所在部位。通过仿真、实验和现场信号的分析,验证了可通过拟合故障振动信号模型实现故障部位的准确诊断。In order to overcome the shortcomings of traditional fauh diagnosis, the fault diagnosis of rolling bearing based on the data fitting of fault vibration signal model is proposed, its validity and practicability are verifwd by the gearbox of wind turbine. The fault vibration signal model is present according to the characteristics of vibration signal, and then data fitting is carded out on the model through the genetic algorithm, the data is the IMF components which are the result of original vibration signal decomposition by EMD, the damage is located by comparing the fault characteristic frequency and the fitting results. The simulation, the experimental and the actual signals are analyzed; the results demonstrate that fault diagnosis can be realized by fitting the fault vibration signal model.

关 键 词：滚动轴承 故障信号模型 数据拟合 EMD 遗传算法 

分 类 号：TH16[机械工程—机械制造及自动化] TH133.33