摘要
为了克服气液二相流特征融合后不相关特征过多的问题,提出了基于距离评估和支持向量机(SVM)的气液二相流流型识别方法。首先利用经验模式分解和小波包方法对原始的压差波动信号进行分解,分别提取原始信号和各分解信号的时域特征参数组成融合特征,然后采用距离评估方法对融合特征进行评估,根据距离评估因子的大小挑选出敏感特征作为SVM的输入,进而实现对流型的自动识别。水平管内空气-水二相流流型识别结果表明:该方法能够准确获取流型的敏感特征,减小运算规模,提高识别准确率。
In order to decrease the irrelevant features of gas-liquid two-phase flow after being fused,a novel identification method of gas-liquid two-phase flow regimes based on distance evaluation and support vector machine(SVM) was proposed.The differential pressure fluctuation signals were decomposed via empirical mode decomposition(EMD) and wavelet packet method respectively.The characteristic parameters in time domain were extracted from the original signals and each decomposed signal to construct the fusion features.Furthermore,a feature evaluation method was applied to calculate the evaluation factors of the fusion features,and the corresponding sensitive features were selected according to the evaluation factors,and then they were input into the SVM to automatically identify flow regime.The identification results of air-water two-phase flow regime in horizontal pipe show that this method enables to precisely extract flow regime sensitive feature,reduce the scale of operation,increase the identification accuracy.
出处
《化学工程》
CAS
CSCD
北大核心
2010年第9期19-22,共4页
Chemical Engineering(China)
基金
国家自然科学基金资助项目(50706006)
关键词
特征选择
经验模式分解
小波包
距离评估
流型识别
feature selection
empirical model decomposition
wavelet packet
distance evaluation
flow regime identification