滚动轴承振动诊断的SOM神经网络方法  

Vibrating diagnosis of rolling bearings based on self-organizing feature map neural network

作　　者：吴涛 原思聪[2] 孟欣[2] 张满意[2] 刘道华[2] 

机构地区：[1]中煤西安设计工程有限责任公司,西安710054 [2]西安建筑科技大学机电工程学院,西安710055

出　　处：《机械设计与制造》2010年第1期198-200,共3页Machinery Design & Manufacture

基　　金：陕西省自然科学基金(2007E_218);西安建筑科技大学科技基金资助(JC0711)

摘　　要：归纳和总结了SOM神经网络多参数诊断法的实施步骤,阐述了轴承故障与振动信号之间的关系以及SOM神经网络的工作原理和实现过程,通过试验研究,提取了反映滚动轴承故障类型的振动信号的特征参数,以构建训练神经网络的特征向量,利用MATLAB 7.0人工神经网络工具箱(ANN)模拟和仿真SOM神经网络,然后用训练后的SOM神经网络对故障模式进行识别。It summarizes the steps of the multi-parameter diagnosis method based on Self-Organiz- ing Feature Map neural network and illuminates the connection between fault and vibration signal of rolling bearings, and operating principle and implementation procedure of SOM neural network. Via experi- mental investigation,the characteristic parameters which are able to image the fault type of the rolling bearing are extracted to construct the characteristic vectors of training a neural network. The SOM neural network is simulated using ANN （Artificial Neural Network）toolbox of MATLAB7.0 and then the fault mode is identified with the trained SOM neural network.

关 键 词：振动 滚动轴承 故障诊断 SOM神经网络 

分 类 号：TH16[机械工程—机械制造及自动化] TH133.31