摘要
针对气液二相流压差波动信号的非平稳和非线性特征,提出了一种基于经验模式分解(EMD)复杂度特征和支持向量机的流型识别方法。该方法首先对二相流压差波动信号进行经验模式分解,将其分解为若干个固有模态函数(IMF),然后对每一个IMF分量提取复杂度特征作为流型特征向量,并以此作为输入参数建立支持向量机分类器来识别流型。对水平管内空气-水二相流的实验结果表明,文中提出的方法能准确地识别流型,从而为流型识别提供了一种新的有效方法。
Aiming at the non-stationary and nonlinear characteristics of differential pressure fluctuation signals of gas-liquid two-phase flow, flow regime identification method based on complexity characteristic and support vector machine was put forward. Empirical Mode Decomposition (EMD) method was used to decompose the differential pressure fluctuation signals into a finite number of stationary Intrinsic Mode Functions (IMF), then complexity features of each IMF was extracted as the flow regime characteristics vectors and served as input parameters of SVM classifier to identify flow regime. The experimental identification results show that the proposed approach can identify flow regime accurately for identifying four typical flow patterns of air-water two-phase flow in horizontal pipe. This method provides a new effective way to identify flow regime.
出处
《化学工程》
EI
CAS
CSCD
北大核心
2008年第4期27-30,共4页
Chemical Engineering(China)
基金
吉林省教育厅重点项目(2006024)
关键词
流型识别
经验模式分解
复杂度
支持向量机
flow pattern identification
Empirical Mode Decomposition
complexity
Support Vector Machine