基于Chebyshev最佳逼近的偏正交分解特征提取方法  

Feature extraction based on best uniform approximation of Chebyshev polynomial with nonlinear partial orthogonal decomposition

作　　者：柳桂国[1] 楼望建 黄道[1] 

机构地区：[1]华东理工大学工业自动化国家工程中心分部 [2]宁波华盛科教仪器有限公司,宁波315040

出　　处：《仪器仪表学报》2008年第6期1275-1279,共5页Chinese Journal of Scientific Instrument

摘　　要：本文提出一种新的非平稳非线性信号特征提取方法。首先,利用Chebyshev插值法建立非平稳信号的最佳一致逼近函数;然后,通过对该函数进行偏正交分解获取对应的特征值及特征向量。特征值及特征向量作为信号特征,具有较高的稳定性。该方法计算量少,具有较高的模式稳定性。运用该方法通过对CWRU(Case Western Reserve University Bearing Data Center)轴承数据分析计算,得到正常滚动轴承与各类故障滚动轴承的特征值向量。实验表明:特征值向量与正常滚动轴承,故障滚动轴承有明显的差异,取得了良好的分类效果。A novel feature extraction method for nonlinear and nonstationary signal is put forward in the paper. Firstly, the best uniform approximation function is constructed with Chebyshev interpolation method, then the function is analyzed by partial orthogonal decomposition to obtain related eignvalue and eignvalue vector. The vector reflects the pattern of the vibration signal steadily. The data from CWRU （ the Case Western Reserve University Beating Data Center） were analyzed using the proposed method, and the result shows that the patterns for vibration signals of normal and different kinds of fault rolling bearings are extracted steadily and effectively.

关 键 词：逼近函数 CHEBYSHEV多项式 偏正交分解 特征提取 滚动轴承 故障诊断 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]