基于ICA和SVM的滚动轴承故障诊断方法研究

A fault diagnosis approach for roller bearing based on ICA and SVM

通过对滚动轴承振动信号的分析处理,提出了基于独立分量分析和支持向量机的故障诊断方法,采用FastICA算法对信号进行分析处理,提取出代表轴承运行状态的投影系数矩阵,并以此作为特征向量来建立支持向量机分类器,利用SVM网络的智能性来判断滚动轴承的工作状态和故障类型。

By analyzing and processing the vibration signals of roller bearing, a fault diagnosis approach based on ICA and SVM is presented. The projection coefficients matrix which represent operating state of the bearing are extracted by analyzing and disposing the signal of bearing on the FastICA method, then taking the matrix as characteristic vectors to definite the fault classifier of the support vector machine. The condition and fault pattern of the roller bearing can be identified with the intellectual ability of SVM network.