动平衡信号的时频分析方法研究

针对动平衡信号面向去重平衡机应用时存在的非线性、非平稳性以及实时性,本文在分析硬支撑动平衡理论的基础上,以联合时频分析方法对标准工件进行加重获得的动平衡信号为研究对象,分别用短时傅立叶变换(STFT)、Wigner-Ville分布(WVD)、小波变换和Hilbert-Huang变换(HHT)对动平衡信号进行时频分析。仿真结果表明,STFT是基于窗函数变换的分析方法,对信号局部分析能力不足且分析精度不高;WVD虽然具有好的时频聚集性,但对多分量信号存在交叉干扰项;小波变换局部分析能力强且分析精度高,但涉及小波基的选择,对非平稳信号不具有实时性;HHT可根据信号的局部时变特征对其进行自适应分解,最终给出信号的时频谱和幅值谱,能够准确描述动平衡信号的特征,具有很高的时频分辨率和时频聚集性。

Time-frequency Analysis for Dynamic Balancing Signal

The balancing machine has the problems which include nonlinear, non-stationary and real-time. This paper analyzes dynamic balancing signals time-frequency characteristics based on support balancing theory, which uses short-time Fourier Transform (STFT),Wigner-Ville distribution (WVD),wavelet transform and Hilbert-Huang trans-form (HHT). And dynamic balancing signals obtained by aggravating a joint standard work piece were taken as a re-search object. The simulation results show that STFT on signal local capacity is insufficient and the responding accura-cy is low. Although WVD has a good character of spatial-frequency concentration, multi-component signal crosstalk terms exist. Wavelet transformation has good capacity on signal local and high analysis accuracy,but it is not real-time due to the choice of the wavelet basis. The signal can be adaptively decomposed by HHT according to their local char-acteristics, resulting in their spectrum and amplitude spectrum. The method can accurately characterize dynamic balanc-ing signals. It has high time-frequency resolution and time-frequency concentration.