基于多尺度熵分析的CO_(2) 气液两相流流型识别  

Flow Pattern Identification of CO_(2)Gas-liquid Two-phase Flow Based on Multiscale Entropy Analysis

作　　者：张文彪[1] 王港华 邵丁 章杰 ZHANG Wenbiao;WANG Ganghua;SHAO Ding;ZHANG Jie(School of Control and Computer Engineering,North China Electric Power University,Beijing 102206,China)

机构地区：[1]华北电力大学控制与计算机工程学院,北京102206

出　　处：《计量学报》2024年第7期1024-1030,共7页Acta Metrologica Sinica

基　　金：国家自然科学基金(61973113,62073135)。

摘　　要：利用四电极对壁式电容传感器中对流型变化最敏感的一组极板对,获取CO_(2)气液两相流不同流型的电容时间序列,并采用多尺度熵算法进行分析。依据多尺度熵曲线的3个特征:前端多尺度熵率、中段多尺度熵率和末端多尺度熵截距,对不同气液两相流工况进行流型识别,并将识别结果与高速摄像机获得的两相流照片判断的流型进行对比。研究表明,通过电容时间序列多尺度熵曲线能够充分展示两相流动的动力学特征,使用多尺度熵曲线的3个特征进行流型识别,能够较为准确地识别分层流、段塞流、泡状流和混状流4种流型,识别准确率优于99%。A group of electrode pairs that are most sensitive to the change of flow patterns are selected from four-electrode concave sensor to obtain capacitance time series from different flow patterns of CO_(2)gas-liquid two-phase flow and capacitance time series are analyzed with multi-scale entropy(MSE).Three characteristics in the multiscale entropy curve,which are front multiscale entropy rate,middle multiscale entropy rate,terminal multiscale entropy intercept,are used to identify the flow pattern under different gas-liquid two-phase flow conditions.The identification results are compared with the real flow pattern from two-phase flow pictures obtained by the high-speed camera.The results show that the dynamic characteristics of two-phase flow can be fully represented by the multiscale entropy curve of the capacitance time series and the four flow patterns of stratified flow,slug flow,bubble flow and churn flow can be identified using the three characteristics of the multiscale entropy curve with the accuracy of more than 99%.

关 键 词：气液两相流 二氧化碳 碳捕集与封存 电容传感器 多尺度熵 流型识别 

分 类 号：TB937[一般工业技术—计量学]