摘要
为了去除气液两相流差压信号中的噪声成分,并提取差压波动信号,提出了一种基于Hilbert-Huang Transform(HHT)的气液两相流信号处理方法.对不同流型下的差压信号进行固有模态分解,分析不同模态的边际谱,利用HHT的多尺度滤波特性除去噪声,进一步得到差压波动信号.实验结果表明,该方法可以有效去除高频噪声,得到的差压波动信号又很好地保持了信号的局部特性.该方法原理简练、物理意义明确,是处理非线性信号的优良方法.
For wiping off noise component from differential pressure signal of gas-liquid two-phase flow, and extracting differential pressure fluctuation signal, a signal process method of gas-liquid two-phase flow based on HHT （Hilbert-Huang Transform） was put forward. Differential pressure signal of different flow regime were handled by EMD （Empirical Mode Decomposition）, marginal spectrum of different mode were analyzed, and noise were eliminated using multi-dimension filter characteristics of HHT. Further, differential pressure fluctuation signal were attained. Experimental results show that this method can get rid of high frequency noise effectively, acquired differential pressure fluctuation signal can commendably keep the partial characteristics of signal. Clear principle and concise physical meanings make it be a fine method to process nonlinear signal.
出处
《传感技术学报》
CAS
CSCD
北大核心
2007年第4期862-865,共4页
Chinese Journal of Sensors and Actuators
基金
国家质量监督检验检疫总局规划项目资助(2006QK23)
国家高技术研究发展计划专项基金资助(2001AA413210)
中国计量学院校立项目联合资助(XZ0516)
关键词
两相流
差压波动信号
多尺度滤波
HHT
two-phase flow
differential pressure fluctuation signal
multiple-scale filter
Hilbert-Huang Transform
作者简介
孙斌（1972-），1999年7月毕业于哈尔滨工业大学电力传动及其自动化专业，获硕士学位．2002年-2005年就读于浙江大学控制科学与工程专业，2005年3月获博士学位．2006年11月在浙江大学信息学院博士后流动站工作，副研究员．研究方向为多相流参数检测、流量仪表设计、远程测控系统设计等，bsun555@cjlu．edu．cn
张宏建，博士，浙江大学教授、博士生导师．
