基于多层加权复杂网络的气液两相流流型分析  

Gas-liquid Two-Phase Flow Pattern Analysis Based on Multilayer Weighted Complex Network

作　　者：张立峰[1] 王智 张启亮 ZHANG Li-feng;WANG Zhi;ZHANG Qi-liang(Department of Automation,North China Electric Power University,Baoding,Hebei 071003,China)

机构地区：[1]华北电力大学自动化系,河北保定071003

出　　处：《计量学报》2023年第5期735-742,共8页Acta Metrologica Sinica

基　　金：国家自然科学基金(61973115)。

摘　　要：提出一种基于多层加权复杂网络的流型分析方法。首先利用电阻层析成像系统获取垂直上升管道气液两相流流动信息,并将测量数据压缩处理以简化数据分析,然后使用多元经验模态分解算法对其进行多尺度分解,进而将流动系统映射到多层加权网络中,通过计算平均加权聚集系数与谱半径定量描述网络结构。研究结果表明,该网络模型可有效揭示泡状流到段塞流的演化过程,从气泡的聚合发展到气塞的逐渐破碎,从伪周期性的出现到衰退都可被网络参数的变化所反映。A flow pattern analysis method based on multi-layer weighted complex networks is presented.Firstly,the electrical resistance tomography system is used to obtain the flow information of gas-liquid two-phase flow in vertical rising pipeline,and the measured data are compressed to simplify the data analysis.Then the multi-scale decomposition is carried out by using the multi-dimensional empirical mode decomposition algorithm,so the flow system can be mapped to the multi-layer weighted network.The network structure is described quantitatively by calculating the average weighted aggregation coefficient and spectral radius.The final results show that the network model can effectively reveal the evolution process from bubble flow to slug flow,from bubble aggregation to gas slug gradual breaking,and from pseudo periodicity to decline can be reflected by the changes of network parameters.

关 键 词：计量学 气液两相流 电阻层析成像 多层加权网络 多元经验模态分解 

分 类 号：TB937[一般工业技术—计量学]