【摘要】 提出了基于小波去噪、降采样和HHT变换的方法。该方法先利用小波进行信号去噪,克服噪声对EMD分解的影响。其次,为获得正确的IMF分量和Hilbert谱,采用降采样方法对信号进行重采样,继而得到适当的采样率。最后,进行EMD分解提取具有明确物理意义的水轮机振动模式分量信号,再对各分量信号进行Hilbert谱分析,从而识别信号的异常频率和发生时间。并将该方法应用于某电站1号机组振动信号分析,结果表明,基于小波预处理的水轮机振动信号Hilbert-Huang变换方法能对机组性能做出良好评价,值得推广应用。更多还原

【Abstract】 This paper put forward a method which is based on wavelet pretreatment and downsampling combined with Hilbert-Huang Transform(HHT).First,this method requires the wavelet de-noising of filtered signals to overcome the noise influence on HHT.Besides,to achieve correct IMF component and Hilbert spectrum,the downsampling method is used to resample the signals so as to obtain appropriate sampling rate.In the end,EMD is decomposed to extract the hydraulic turbine vibration mode component signals with explicit physical meaning and each component signal is going through Hilbert Spectrum analysis,so that the abnormal frequency and occurrence time of the signals can be identified.This method was applied to a power station on the 1st vibration signal analysis,the results show that the pretreatment of turbine vibration signal wavelet Hilbert-Huang transform method can make a good evaluation of performance on the unit,should be widely applied.更多还原