管束间气液两相流绕流压差波动信号的复杂性及混沌形态分析

基于两种节距比管束间的不同流型的压差波动信号,对其进行复杂性测度特征值提取(盒子维数和近似熵),分析了两种复杂性测度随气相折算速度变化的动力学特性,并研究了典型流型下压差波动信号的混沌吸引子形态特征表征气液两相流流型的能力。结果表明:两种复杂性测度的变化趋势及典型流型下的吸引子形态反映出它们对流型的变化是敏感的。可以表征两相流动力学结构变化,有益于两相流动力学机理的研究,为揭示两相流流型演化机理和定量标识流型提供了一种有效的辅助工具。     

油水两相流相含率的软测量方法

对垂直上升管中油水两相流流动测得的电导波动信号,从时频域内提取了11个反映油水两相流流动特性的特征量.频域内特征量提取采用了语音信号处理中的线性预测方法,时域内特征量提取采用了时间序列统计分析方法.将这些特征量作为人工神经网络的输入量,在总流量10～60 m³•d^-1及含水率51%～91%范围内,采用基于Levenbery-Marquardt算法的BP人工神经网络作为相含率预估模型,较好地实现了油水两相流含水率预测,为两相流相含率测量提供了一种新的软测量途径.     

两相流流动结构多尺度复杂熵因果关系平面特征

采用一种新的信息熵理论分析了气液两相流流动结构动力学特征。首先考察了典型混沌时间序列的多尺度复杂熵因果关系平面特征,然后将该方法应用于气液两相流3种典型流型(泡状流、段塞流及混状流)的电导传感器波动信号处理,结果表明单尺度复杂熵因果关系平面能够对不同流型线性识别。此外发现多尺度复杂熵因果关系平面能够描述流动结构信息随尺度增加的连续丢失过程,反映了气液两相流流动结构稳定性及复杂性,是理解两相流流动结构非线性动力学特性的一种有用分析工具。     

沸腾管内汽液两相流动压力波动信号的分析

研究了竖直管内沸腾两相流动时压力波动信号的特性,采用功率谱密度函数、分维数、关联维数及Kolmogorov熵等非线性分析方法对压力波动时间序列进行了定性及定量分析,考察了热流密度及液速对压力波动的影响规律。结果表明,此类系统具有非线性混沌特性。定量计算结果表明,分维数小于1.5,系统存在1个关联维。K熵的变化也具有同样的趋势,且为有限正值;沸腾流动的汽液两相流型和非线性特征量之间有一定的关系。     

2种熵测度在分析流型信号复杂度上的应用

为研究气液二相流动特性,更深地了解气液二相流动,文章利用非线性复杂性测度:近似熵和样本熵,研究了气液二相流的3种流型,泡状流、段塞流和雾状流流型波动信号的复杂度,并对影响算法的容差r和序列长度N进行了讨论。结果表明:泡状流的熵值最高,段塞流的熵值最低,雾状流的熵值居中,且当序列长度N大于400时,近似熵和样本熵可以很好地识别泡状流、段塞流、雾状流。从计算结果可以看出,泡状流的复杂性最高,雾状流其次,段塞流最低,这说明泡状流的流动情况最为复杂,段塞流的流动最为规律。泡状流随机可变特性表现为熵的高值及振荡特征,段塞流气塞与液塞的间歇性运动表现为熵的低值及平稳性,雾状流极不稳定的振荡运动特性表现为介于泡状流及段塞流之间的熵值特点。     

基于Hilbert-Huang变换和支持向量机的油水两相流流型识别

针对油水两相流的测量难题，利用文丘里管对水平管内油水两相流流型进行了研究。基于差压波动信号，提出了Hilbert-Huang变换与支持向量机相结合的流型识别方法。首先计算差压波动信号的均方根，并对其进行归一化处理后作为表征流型的特征向量之一；然后对差压信号进行Hilbert-Huang变换，利用经验模态分解后的多分辨率特征，提取第一层和第二层的能量比作为表征流型的另外两个特征向量；最后利用训练好的支持向量机进行流型识别，对分层流型及分散流型取得了较好的流型识别效果。如果将流型识别与推导得到的文丘里油水两相流流量测量模型相结合，可以较好地实现油水两相流的流量测量。     

垂直上升管中油水两相流流型表征

在含水率 5 1%～ 91%、油水两相流总流量 10～ 6 0m³·d^-1范围内 ,对垂直上升油水两相流流动工况的电导波动信号进行了分形及混沌时间序列分析 .当含水率为 6 0 .5 %～ 91%时 ,所提取的分形维数及吸引子相关维分别在 1.0 0 6～ 1.6 31及 4.30～ 6 .77之间 ;当含水率为 5 1%或 5 1.5 %时 ,所提取的分形维数及吸引子相关维分别在 1.17～ 1.75 8及 5 .46～ 7.0 5之间 ,且分形维数及混沌吸引子相关维数随总流量呈不规则突变 ,并与复杂功率谱特征及流型图上的过渡流型相对应 ,表明了分形维数及混沌吸引子相关维数对油水两相流流型变化具有敏感的“指示器”特性     

油水两相流流型混沌吸引子形态特性

基于在垂直上升管中采集到的61组油水两相流电导波动信号,通过在混沌吸引子域中放置参考截面方法,考察了混沌吸引子形态特征量与油水两相流流型转变之间的关系,发现水包油流型在3阶及4阶吸引子矩特征量平面上分布相对稳定,而过渡流型在3阶吸引子矩特征量平面上呈不规则剧烈变化,在4阶吸引子矩特征量平面上呈线性变化。研究结果表明,油水两相流混沌吸引子形态特征量对流型转化敏感,有助于理解油水两相流流型转化的非线性动力学机制。     

混沌递归分析在油水两相流流型识别中的应用

采用由Lorenz方程生成的时间序列验证了递归定量分析的有效性及敏感性,并将该分析方法应用于油水两相流流型识别.研究结果表明：对油水两相流过渡流型,在递归图上表现为分散点或线条纹理两种纹理结构特征;而对同一种水包油流型,仅表现为线条纹理结构特征.表明混沌递归图纹理结构特征对油水两相流流型变化敏感,混沌递归定量分析是一种流型识别的有效辅助诊断工具.     

水平气液两相流流型空间图像信息复杂性测度分析

为了考察从图像灰度序列提取的复杂性测度与气液两相流流型变化之间的关系,本文首先从高速摄影系统拍摄的60种流动工况下水平管内气液两相流流型图像中提取了三种复杂性测度（Lempel-Ziv复杂度,分形盒维数,Shannon信息熵）,在此基础上研究了不同表观气速下三种复杂性测度的混沌动力学特性,以及对气液两相流流型的表征能力。实验结果表明：三种复杂性测度均能敏感地指示出流型的变化;通过对三种复杂度随两相流流动参数变化规律分析,可以得到气液两相流动力学结构反演特征, 为揭示气液两相流流型转化机理和定量识别流型提供了一种有效的辅助诊断工具。     

利用符号时间序列分析法表征垂直上升管中油水两相流流型

The flow patterns of oil/water two-phase flow in vertical upward pipes were characterized by the analysis of symbolic time series based on the conductance fluctuating signals. The study showed that the symbolic sequence temporal irreversibility Tfb and chi-square χfb^2 statistics had little change with oil-in-water flow pattern variations for water cut (Kw) ranging from 61% to 91% but showed irregular sudden changes with transitional flow pattern variations for water cut 51%. When distinguishing the transitional flow pattern from oil-in-water flow pattern, the symbolic time series analysis method presented more unique characteristics and was a useful assistant diagnostic tool for the identification of oil/water two- phase flow patterns.     

汽液固三相流动沸腾传热系统的非线性特性

运用功率谱分析、自相关分析、相图、关联维数和K熵等非线性分析技术对汽液固三相流动沸腾传热系统中的温度波动信号进行了定性和定量研究.研究表明：此类系统具有非线性混沌特性;定量计算结果表明,系统存在2～3个分维,K熵的变化也具有同样的趋势,且为有限正值;传热特性和非线性特征量之间有一定的关系,即颗粒体积含量越大,非线性特征量越大,传热系数越大.     

Experimental Investigation on the Holdup Distribution of Oil‐Water Two‐Phase Flow in Horizontal Parallel Tubes

An experimental investigation was conducted to study the holdup distribution of oil and water two‐phase flow in two parallel tubes with unequal tube diameter. Tests were performed using white oil (of viscosity 52 mPa s and density 860 kg/m³) and tap water as liquid phases at room temperature and atmospheric outlet pressure. Measurements were taken of water flow rates from 0.5 to 12.5 m³/h and input oil volume fractions from 3 to 94 %. Results showed that there were different flow pattern maps between the run and bypass tubes when oil‐water two‐phase flow is found in the parallel tubes. At low input fluid flow rates, a large deviation could be found on the average oil holdup between the bypass and the run tubes. However, with increased input oil fraction at constant water flow rate, the holdup at the bypass tube became close to that at the run tube. Furthermore, experimental data showed that there was no significant variation in flow pattern and holdup between the run and main tubes. In order to calculate the holdup in the form of segregated flow, the drift flux model has been used here.     

Nonlinear time‐series analysis of optical signals to identify multiphase flow behavior in microchannels

This study provides an insight into the instability and irregularity of multiphase flows in microchannels. Using a homemade optical measuring system, time series related to two‐phase flow dynamics under different operating conditions, fluids, and channel lengths were collected and analyzed via a nonlinear characteristic parameter, Kolmogorov entropy (K_E). Our results reveal that higher K_E corresponds to unstable flow behavior, while lower K_E refers to steady flow behavior; higher K_E values appear at comparatively low or high gas flow rates, and most Taylor flow regime appeared at proper operating conditions with small K_E. An equation derived based on the definition of K_E is proposed to better understand K_E characteristics, which include bubble break‐up impact and gas/liquid flow rate ratio. Predictions from the proposed analytical equation agree with experimental results, suggesting that the equation effectively identifies unstable flows and can be used to ensure stable and predictable multiphase flows in microchannels. © 2016 American Institute of Chemical Engineers AIChE J, 63: 2378–2385, 2017     

大深宽比双T形微通道内液-液两相流可视化研究

通过高速摄像对水力直径0.176mm、深宽比2.4的双T形矩形微通道内的液-液两相流动进行了可视化实验研究。改变连续相（硅油）和分散相（水）的流量比,记录分析了微通道不同部位油-水两相流的流型和流型发展演化情况。实验结果表明,在微通道上游T形部位的油-水两相流型主要包括滴状流、弹状流、波平行流和平行流;在微通道的中间部位,绘制了基于水和硅油量纲为1韦伯数的流型图,并将其与相关文献进行了比较。同时,发现微通道内液塞及液滴的长度（量纲为1）与油/水流量比之间存在线性关系,液塞/液滴速度比两相混合物表观速度大,建立了能够准确描述液塞/液滴运动速度的实验拟合公式。最后,研究了液滴在微通道下游T形部位的行为,观察到断裂和不断裂两种模式并进行了分析,给出了划分断裂与否的流型图。     

Chaotic characteristics of local voidage fluctuation in a circulating fluidized bed

Local voidage fluctuations have been measured by using an optical transmittance probe at various axial and radial positions in a circulating fluidized bed riser with a 0.1 m i. d. and 10 m height. The chaotic time series analysis of the local voidage fluctuations has been adopted to characterize the nonlinear dynamics of the circulating fluidized bed riser. The variations of the correlation dimension and the Kolmogorov entropy of the voidage fluctuation were found to depend on the local time-average voidage. The axial and radial distributions of the correlation dimension and the Kolmogorov entropy were strongly affected by the solids flow structures (e.g. core-annulus flow) in various operating conditions. The correlation dimension of local voidage fluctuations increases along the riser, except the position near the distributor. Both, the correlation dimension and the Kolmogorov entropy of local voidage fluctuations near the wall, were found to be smaller than those at the center of the riser, independent of the solids circulation rate and the axial position.     

流体物性对壳侧不互溶双组分两相流动特性的影响

选用空气-水和空气-柴油2种不互溶双组分两相混合物物系,实验研究了液相物性对管壳式换热器壳侧两相流流型及其转变和两相流动特性的影响.研究表明,液相物性对间歇流向环状流的转变有显著影响,对于向泡状流转变的影响则不如在管内流动明显.研究还表明,液相物性特别是粘度和表面张力对截面含气率和两相摩擦压降有显著影响.     

Very-viscous-oil/water/air flow through horizontal pipes: Pressure drop measurement and prediction

Core annular flow pattern, where a low viscosity liquid surrounds a very-viscous one, may be very interesting for heavy oil transportation. However, in oil production, oil and water rarely flow alone and gas is usually present. Despite several publications on liquid-liquid core annular flow, no much work has been done towards a proper characterization of the effect of gas on pressure drop. The aim of this paper is twofold: to provide a new data base on three-phase (very-viscous-oil/water/air) flow, and to propose a simple model for the determination of pressure drop.     

Flow Regime Identification in a Bubble Column via Nuclear Gauge Densitometry and Chaos Analysis

The bubble column performance can change significantly as a result of flow regime change. Since reactor volume productivity, mass and heat transfer as well as mixing are affected by the prevailing flow regime, it is very important to know how to identify it. In this work, flow regime identification was performed on the basis of the Kolmogorov entropy (KE) algorithm applied to nuclear gauge densitometry data. In addition, the average cycle time was used for validation of the results. Three transition velocities were identified that delineated the boundaries of the three main hydrodynamic regimes. The first two transition points were also confirmed by the information entropy concept. The increasing KE trend in the bubbly flow regime and the decreasing KE trend in the churn‐turbulent regime were predicted successfully by means of new semi‐theoretical models.