基于改进的局部特征尺度分解和归一化正交的时频分析方法

针对希尔伯特-黄变换(Hilbert-Huang Transform,HHT)中经验模态分解(Empirical Mode Decomposition,EMD)和希尔伯特变换(Hilbert Transform,HT)的不足,提出了一种基于改进的局部特征尺度分解(Improved Local Characteristic-scale Decomposition,ILCD)和归一化正交(Normalized Quadrature,NQ)的时频分析方法.ILCD克服了LCD的固有缺陷,在精确性和正交性等方面要优于EMD和LCD方法.同时为了克服HT的不足,提出了一种基于经验调幅调频分解标准化估计瞬时频率的归一化正交(Normalized Quadrature,NQ)方法.通过仿真信号将EMD,LCD与ILCD进行对比,同时将标准希尔伯特变换,直接正交法与NQ进行对比,结果表明了论文方法的有效性.转子碰摩故障数据分析结果进一步验证了论文方法的有效性和优越性.

A New Time-Frequency Analysis Method Based on Improved Local Characteristic-Scale Decomposition and Normalized Quadrature

Aiming to resolve the problems of Hilbert-Huang transform(HHT),which contains empirical mode decomposition(EMD)and Hilbert transform(HT),a new time-frequency analysis method based on improved local characteristic-scale decomposition(ILCD)and normalized quadrature(NQ)was proposed in this paper.ILCD solved the intrinsic problem of LCD and outperformed the EMD and LCD in accuracy and orthogonality.In addition,the NQ method for estimating instantaneous frequency was proposed based on the empirical AM-FM decomposition.By comparing ILCD with EMD and LCD,and comparing NQ with normalized HT and direct quadrature(DQ),the analysis results indicate that the proposed method is predominant.Finally,the proposed method was applied to fault diagnosis of rotor local rubbing and the conclusions verified its effectiveness and excellence.