摘要
为分析气液两相流差压信号的内在特征,提高测量信号的信噪比(SNR),提出了一种基于小波变换和高阶统计量的信号分析方法.该方法利用小波变换技术将差压信号分解为多个不同频段的尺度函数,根据先验知识滤除信号中的高频噪声,同时提取不同尺度上的细节信号能量特征值.基于高阶统计量技术,提取不同工况下重构的已除噪差压信号的双谱特征值,分析了信号的双谱特性.初步研究表明,小波变换和高阶统计量相结合的信号分析方法能有效地抑制信号中的高斯有色噪声,提取的细节尺度能量特征和双谱特征可以提供更多的关于管道内复杂流动状况的信息.该方法为气液两相流流动状态的判别和过程监控提供了有益的借鉴.
A novel signal analytical method based on wavelet transform and higher-order statistics was presented for analyzing the intrinsic characteristics of differential pressure signal of gas-liquid two phase flow and enhance the signal to noise ratio（SNR）. First, the differential pressure signal was decomposed into many scale functions with different frequencies by using wavelet multi resolution technique. The high-frequency noise of the signal was eliminated according to prior experiences. Meanwhile, the energy eigenvalue of detailed signal was obtained. Under different operational states, the bispeetral eigenvalues of the reconstructed denoising signal was extracted, and the bispectral characteristics were analyzed. Preliminary study results show that the proposed method can effectively suppress Gaussian noise of signal, and that the combination of the scale energy and bispeetral eigenvalue can provide more useful information, which makes this method potentially useful for flow state identification and process monitoring.
出处
《浙江大学学报（工学版）》
EI
CAS
CSCD
北大核心
2006年第4期714-718,共5页
Journal of Zhejiang University：Engineering Science
基金
国家自然科学基金资助项目(50576084)
中国博士后科学基金资助项目(2005038631)
关键词
小波变换
高阶统计量
双谱
差压信号
wavelet transform
higher-order statistics
bispectrum
differential pressure signal
