基于谱图和约束NMF的滚动轴承特征频谱提取算法

Feature Spectrum Extraction Algorithm for Rolling Bearings Based on Spectrogram and Constrained NMF

为了克服传统时频分析中信号特征频谱提取技术中参数敏感问题,设计了一种基于谱图和约束NMF的特征频谱提取算法。该算法首先对振动信号进行归一化预处理和短时Fourier分析,获得代表原始非平稳信号特性的瞬时参数即谱图;然后对谱图进行约束NMF分解;最后由基矩阵获得原始振动信号的特征频谱。理论分析、仿真和工程试验验证了该算法的可行性和有效性。

In order to solve the problem about sensitive parameters in signal feature spectrum extraction of traditional time-frequency analysis,a feature spectrum extraction algorithm is proposed based on spectrogram and Constrained Non-Negative Matrix Factorizations（CNMF）.Firstly,the vibration signal is normalized,and then the short time Fou-rier analysis is carried out,obtaining the instantaneous parameters that represent original non-stationary signal charac-teristics.Secondly,the spectrogram is decomposed using CNMF.Finally,the feature spectrum of original vibration sig-nal is got by the basic matrix.The feasibility and effectively of the proposed method are verified by theoretical analysis, simulation and engineering experiment.