用HHT变换处理离心压缩机喘振试验数据

Centrifugal Compressor Surge Data Processing Based on HHT Transforms

为了提取离心压缩机早期喘振特征频率,在对信号进行小波包降噪抽样后,利用Hilbert-Huang变换(HHT)进行信号特征提取。通过经验模态分解(EMD)得到若干固有模态函数(IMF),然后利用相关系数法对IMF进行筛选。通过趋势项和原始信号对比可知压缩机流量减少是造成振动的主因,最后对有效IMF信号进行Hilbert变换,并求其边际谱,提取压缩机喘振频率为7.3Hz。

To extract centrifugal compressor early surge characteristic frequency,the Hilbert-Huang transform（HHT） method is applied to extract signal character after wavelet packet denoising and signal sampling.The empirical mode decomposition（EMD） was used to decompose the signal into a number of intrinsic mode functions（IMFs）.Using the correlation coefficient of the IMF and the undecomposed signal filters the useful component.The vibration of centrifugal compressor is mainly caused by decreasing flow through comparing original signal to trend.Finally,the effective items are transformed to Hilbert marginal spectrum and extracted surge frequency is 7.3Hz.