摘要
针对传统功率谱熵只能刻画总体系统结构复杂性问题,提出既能从宏观角度又能从微观角度反映系统结构复杂性的多频带谱熵的分析方法。分析几种典型信号的多频带谱熵特征,验证多频带谱熵方法的可行性及抗干扰能力;利用环形电导传感器阵列获取动态实验测试数据,计算水平气液两相流波状流、塞状流和弹状流含水率波动的多频带谱熵值,分析3种流型结构复杂性随频带因子变化的演变规律。实验结果表明:波状流的多频带谱熵最低,弹状流多频带谱熵值最高;塞状流的多频带谱熵居于波状流与弹状流之间。3种流型在分析频率为0-8.3 Hz时,呈现近似线性变化的结构复杂性特征以及较强的流型区分度,可以作为流型识别的准则。
Power spectral entropy can not meet basic requirements of the local structural complexity analysis for two phase flow. A new method based on multiband spectral entropy is proposed to study the structural complexity of horizontal gas / liquid two-phase flow. Firstly,examples using time series from classical systems are provided to demonstrate the effectiveness and robustness of our approach. Based on the above studies,the multiband spectral entropies of three flow regimes( wave flow,plug flow and slug flow) are calculated with conductance fluctuating signals from the first frequency band to the 100 th frequency band. Experimental results indicate that slug flow presents higher multiband spectral entropy because of air plug and the liquid slug,and their inclusion of small bubbles and droplets with which slug flow has complex structures. Wave flow with wave characteristic shows the lowest multiband spectral entropy value. However,the intermittence of gas slug and liquid slug of plug flow behaves as the entropy between that of slug flow and wave flow. Furthermore,multiband spectral entropies of three flow regimes at 1 ~ 8. 33 Hz exhibit an approximate linear variation and strong ability of flow pattern identification. Multiband spectral entropy is a good approach for structural complexity analysis of two phase flow and the rate of multiband spectral entropy is a new criterion of flow regime identification.
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2015年第5期1138-1146,共9页
Chinese Journal of Scientific Instrument
基金
国家自然科学基金(51176141,61227006)
天津市科技创新体系及平台建设计划(13TXSYJC40200)项目资助
关键词
气液两相流
电导波动信号
结构复杂性
多频带谱熵
gas / liquid two-phase flow conductance fluctuation signal structural complexity multiband spectral entropy