Development of a Flow Regime Map for a Horizontal Pipe With the Multi-Classification Support Vector Machines

2008 ◽  
Author(s):  
L. M. Tam ◽  
A. J. Ghajar ◽  
H. K. Tam ◽  
S. C. Tam
Keyword(s):  
Heat Transfer ◽  
Experimental Data ◽  
Mixed Convection ◽  
Flow Regime ◽  
Single Phase ◽  
Flow Regimes ◽  
Support Vector ◽  
Flow Maps ◽  
Vector Machines ◽  
Flow Regime Maps

For horizontal circular pipes under uniform wall heat flux boundary condition and three different inlet configurations (re-entrant, square-edged, bell-mouth), Ghajar and Tam (1995) developed flow regime maps for the determination of the boundary between single-phase forced and mixed convection using experimental data of Ghajar and Tam (1994). Based on the ratio of the local peripheral heat transfer coefficient at the top and the bottom, the heat transfer data was classified as either forced or mixed convection among the different flow regimes. The forced-mixed convection boundary was then obtained by empirical correlations. From the flow maps, heat transfer correlations for different flow regimes were recommended. Recently Trafalis et al. (2005) used the Multiclass Support Vector Machines (SVM) method to classify vertical and horizontal two-phase flow regimes in 4 pipes with good accuracy. In this study, the SVM method was applied to the single-phase experimental data of Ghajar and Tam (1994) and new flow regime maps were developed. Five flow regimes (forced turbulent, forced transition, mixed transition, forced laminar, mixed laminar) were identified in the flow maps using Reynolds and Rayleigh numbers as the identifying parameters. The flow regimes on the boundaries of the new maps were represented by the SVM decision functions. The results show that the new flow regime maps for the three types of inlets can classify the forced and mixed convection experimental data in different flow regimes with good accuracy.

Correlations Between Pressure Drop and Two-Phase Flow Regimes in Microchannel Networks

2009 ◽  
Author(s):  
Christian Weinmu¨ller ◽  
Dimos Poulikakos
Keyword(s):  
Experimental Data ◽  
Pressure Drop ◽  
Flow Regime ◽  
Single Phase ◽  
Two Phase Flow ◽  
Flow Regimes ◽  
Phase Flow ◽  
Two Phase ◽  
Research Activities ◽  
Wide Range

Microfluidics has experienced a significant increase in research activities in recent years with a wide range of applications emerging, such as micro heat exchangers, energy conversion devices, microreactors, lab-on-chip devices and micro total chemical analysis systems (μTAS). Efforts to enhance or extend the performance of single phase microfluidic devices are met by two-phase flow systems [1, 2]. Essential for the design and control of microfluidic systems is the understanding of the fluid/hydrodynamic behavior, especially pressure drop correlations. These are well established for single phase flow, however, analytical correlations for two-phase flow only reflect experimentally obtained values within an accuracy of ± 50% [3, 4]. The present study illustrates the effect of two-phase flow regimes on the pressure drop. Experimental measurement data is put into relation of calculated values based on established correlations of Lockhart-Martinelli with Chisholm modifications for macroscopic flows [5, 6] and Mishima-Hibiki modifications for microscale flows [7]. Further, the experimental pressure drop data is superimposed onto two-phase flow maps to identify apparent correlations of pressure drop abnormalities and flow regimes. The experiments were conducted in a square microchannel with a width of 200 μm. Optical access is guaranteed by an anodically bonded glass plate on a MEMS fabricated silicon chip. Superficial velocities range from 0.01 m/s to 1 m/s for the gas flow and from 0.0001 m/s to 1 m/s for the liquid flow with water as liquid feed and CO2 as gas. The analysis of the flow regimes was performed by imaging the distinct flow regimes by laser induced fluorescence microscopy, employing Rhodamine B as the photosensitive dye. The pressure drop was synchronically recorded with a 200 mbar, 2.5 bar and 25 bar differential pressure transmitter and the data was exported via a LabView based software environment, see Figure 1. Figure 2 illustrates the experimentally obtained pressure drop in comparison to the calculated values based on the Lockhard-Martinelli correlation with the Chisholm modification and the Mishima-Hibiki modification. For both cases the predications underestimate the two-phase pressure drop by more than 50%. Nevertheless, the regression of the experimental data has an offset of linear nature. Two-phase flow is assigned to flow regime maps of bubbly, wedging, slug or annular flow defined by superficial gas and liquid velocities. In Figure 3 the pressure drop is plotted as a surface over the corresponding flow regime map. Transition lines indicate a change of flow regimes enclosing an area of an anticline in the pressure data. In the direct comparison between the calculated and the measured values, the two surfaces show a distinct deviation. Especially, the anticline of the experimental data is not explained by the analytical correlations. Figure 4 depicts the findings of Figure 3 at a constant superficial velocity of 0.0232 m/s. The dominant influence of the flow regimes on the pressure drop becomes apparent, especially in the wedging flow regime. The evident deviation of two-phase flow correlations for the pressure drop is based on omitting the influence of the flow regimes. In conclusion, the study reveals a strong divergence of pressure drop measurements in microscale two-phase flow from established correlations of Lockhart-Martinelli and recognized modifications. In reference to [8, 9], an analytical model incorporating the flow regimes and, hence, predicting the precise pressure drop would be of great benefit for hydrodynamic considerations in microfluidics.

Comparison of Single and Two-Phase Models for the Study of Mixed Convection Flows With Nanofluids

2010 ◽  
Author(s):  
Mahmood Akbari ◽  
Amin Behzadmehr ◽  
Nicolas Galanis
Keyword(s):  
Heat Transfer ◽  
Experimental Data ◽  
Mixed Convection ◽  
Single Phase ◽  
Volume Fraction ◽  
Solid Particles ◽  
Two Phase ◽  
Prediction Ability ◽  
Numerical Predictions ◽  
Phase Models

The single phase and three different two phase models (Volume of fluid, Mixture and Eulerian) are used to analyse laminar mixed convection flow of Al2O3-water nanofluids in a horizontal tube, in order to evaluate their prediction ability. The flow is considered steady and developing. The fluid’s physical properties are temperature dependent whereas those of the solid particles are constant. A uniform heat flux is applied at the fluid-solid interface. Two different Reynolds numbers and three different volume fractions have been considered. The governing three-dimensional partial differential equations are elliptical in all directions and coupled. Predicted convective heat transfer coefficients, velocity, and temperature profiles, as well as secondary flow’s velocity vectors and temperature contours are compared at different axial positions. To validate the comparisons and verify the accuracy of the results, the numerical predictions are compared with corresponding experimental data. There are essentially no differences between the predictions of the two-phase models; however their results are significantly different from those of the single-phase approach. Two-phase model results are closer to the experimental data, but they show an unrealistic increase in heat transfer for small changes of the particle volume fraction. Hydrodynamically, the two-phase and single-phase approaches perform almost the same but their thermal predictions are quite different.

Flow Regime Observations in a Vertical Annulus With an Inner Roughened Tube

2008 ◽  
Author(s):  
G. D. Harvel ◽  
B. Torica
Keyword(s):  
Heat Transfer ◽  
Pressure Drop ◽  
Multiphase Flow ◽  
Flow Regime ◽  
Nuclear Power ◽  
Single Phase ◽  
Flow Regimes ◽  
Smooth Tube ◽  
Surface Conditions ◽  
Vertical Annulus

The efficiency and safe performance of a nuclear power plant is often reliant upon a solid understanding of the multiphase flow phenomena that occurs in various components and piping networks throughout the heat transport loops. Depending upon the conditions, various flow regimes can be observed and each of these flow regimes would have a different impact upon the heat transfer and pressure drop of the component or piping that the two-phase mixture is traveling in. Thus, many researchers have performed significant studies to predict flow regime. Over the past several years, the maintenance and inspection efforts at various nuclear power plants have shown that ageing phenomena are occurring including corrosion, cracking, erosion phenomena, and deposition of materials. One of the effects of such phenomena is the roughening of the surface which would impact the nature of fluid behaviour, heat transfer behaviour, and mass transfer at the surface. Relationships for surface conditions and pressure drop/heat transfer are already established for single phase flow conditions albeit with significant uncertainty. Yet, the relationships for multiphase flow have not been established and it is assumed that the surface effects are covered through the single phase component. In this work, experimental studies are performed for a vertical annulus geometry. The apparatus consists of glass tubing to allow visualization of the flow regime inside the tubes. The glass tube material is also a reference case for the smooth tube. Experiments are performed in a bubble column mode for superficial gas velocities up to 0.5 m/s. Observations are taken at various developing lengths and varied water inventories. The results are compared to previous work which shows that the smooth tube results match previous measurements. The inner surface of the tube is modified by forming a thin layer of material over the glass surface. The material is initially moldable and allows for the imprinting of different surface shapes and roughnesses. The material cures in a few hours and is resistant to erosion or dissolution in water. Thus, the surface conditions are changed on the inner tube. The experiments indicate that the flow regime transition from bubbly to slug flow and from slug to churn flow occur at lower superficial gas velocities.

Development of a Heat Transfer Correlation for the Transitional Flow in a Horizontal Tube Using Support Vector Machines

2008 ◽  
Author(s):  
L. M. Tam ◽  
A. J. Ghajar ◽  
H. K. Tam ◽  
S. C. Tam
Keyword(s):  
Heat Transfer ◽  
Experimental Data ◽  
Support Vector Machines ◽  
Least Squares Method ◽  
Horizontal Tube ◽  
Good Method ◽  
Transitional Flow ◽  
Support Vector ◽  
Flux Boundary Condition ◽  
Vector Machines

In this paper the Support Vector Machines (SVM) method is used to correlate the transitional forced and mixed convection experimental data of Ghajar and Tam (1994) that were obtained along a stainless steel horizontal circular tube fitted with re-entrant, square-edged, and bell-mouth inlets under uniform wall heat flux boundary condition. The SVM method has been chosen to further improve the accuracy of the correlations that were developed by Ghajar and his co-workers using the traditional least-squares method (Ghajar and Tam, 1994) and more recently the artificial neural networks (ANN) method (Ghajar et al., 2004). Using the ANN method improved the accuracy of their correlation. However, there are drawbacks associated with ANN method. One of the major problems with the ANN method is that it does not provide a unique correlation due to different coefficient matrices. The SVM method used in this study eliminated the drawbacks associated with the ANN method and provided a unique correlation with comparable accuracy as the ANN method. For the experimental data used, majority of the data points were predicted within 5% deviation. Comparisons were made regarding the accuracy of the developed correlation and its characteristic using SVM and ANN methods. The results showed that SVM is a good method to correlate complex heat transfer data.

Buoyancy Effect on Heat Transfer in 5×5 Rod Bundles

2017 ◽  
Author(s):  
Da Liu ◽  
Fujun Gan ◽  
Chaozhu Zhang ◽  
Hanyang Gu
Keyword(s):  
Heat Transfer ◽  
Experimental Data ◽  
Mixed Convection ◽  
Buoyancy Force ◽  
Reynold Number ◽  
Low Flow ◽  
Buoyancy Effect ◽  
Rod Bundles ◽  
Convection Regime ◽  
Dc Power

Experiments of heat transfer at low flow rate are performed in a 5×5 square arrayed rod bundles. The diameter of the rod is 10mm with a pitch of 13.3mm, length of the test section is about 3 meters. Inlet Reynold number ranges from 2000 to 30000, Bo * ranges from 4×10−6 to 5×10−3. The rods are heated using a DC power, the heat flux ranges from 30 to 300 kW/m2. The experiment is aimed to investigate the buoyancy effect of mixed convection in rod bundles. The experimental data shows that similar with mixed convection in circular channels, buoyancy force has great effect on heat transfer at mixed convection regime in rod bundles. But the buoyancy effect appears at higher Bo* conditions. The spacer effect have also been investigated at both turbulent forced convection regime and mixed convection regime. The reconstruction of heat transfer downstream of spacers is different at different flow regimes, a reasonable explanation was provided.

On the Application of Macro-Pipe Two-Phase Heat Transfer Correlations and Flow Regime Maps to Mini-Channels

2006 ◽  
Author(s):  
Avram Bar-Cohen ◽  
Ilai Sher ◽  
Emil Rahim
Keyword(s):  
Heat Transfer ◽  
Heat Transfer Coefficient ◽  
Transfer Coefficient ◽  
Flow Regime ◽  
Annular Flow ◽  
Negative Slope ◽  
Two Phase ◽  
Heat Transfer Correlations ◽  
Two Phase Heat Transfer ◽  
Flow Regime Maps

The present study is aimed at evaluating the ability of conventional “macro-pipe” correlations and regime transitions to predict the two-phase thermofluid characteristics of mini-channel cold plates. Use is made of the Taitel-Dukler flow regime maps, seven classical heat transfer coefficient correlations and two dryout predictions. The vast majority of the mini-channel two-phase heat-transfer data, taken from the literature, is predicted to fall in the annular regime, in agreement with the reported observations. A characteristic heat transfer coefficient locus has been identified, with a positive slope following the transition from Intermittent to Annular flow and a negative slope following the onset of partial dryout at higher qualities. While the classical two-phase heat transfer correlations are generally capable of providing good agreement with the low-quality annular flow data the quality at which partial dryout occurs and the ensuing heat transfer rates are not predictable by the available macro-pipe correlations.

Computational Fluid Dynamics Modeling of Mixed Convection Flows in Building Enclosures

2013 ◽  
Author(s):  
Alexander Kayne ◽  
Ramesh Agarwal
Keyword(s):  
Heat Transfer ◽  
Fluid Dynamics ◽  
Experimental Data ◽  
Computational Fluid Dynamics ◽  
Mixed Convection ◽  
Turbulence Model ◽  
Numerical Algorithm ◽  
Navier Stokes ◽  
Cfd Simulations ◽  
Flow And Heat Transfer

In recent years Computational Fluid Dynamics (CFD) simulations are increasingly used to model the air circulation and temperature environment inside the rooms of residential and office buildings to gain insight into the relative energy consumptions of various HVAC systems for cooling/heating for climate control and thermal comfort. This requires accurate simulation of turbulent flow and heat transfer for various types of ventilation systems using the Reynolds-Averaged Navier-Stokes (RANS) equations of fluid dynamics. Large Eddy Simulation (LES) or Direct Numerical Simulation (DNS) of Navier-Stokes equations is computationally intensive and expensive for simulations of this kind. As a result, vast majority of CFD simulations employ RANS equations in conjunction with a turbulence model. In order to assess the modeling requirements (mesh, numerical algorithm, turbulence model etc.) for accurate simulations, it is critical to validate the calculations against the experimental data. For this purpose, we use three well known benchmark validation cases, one for natural convection in 2D closed vertical cavity, second for forced convection in a 2D rectangular cavity and the third for mixed convection in a 2D square cavity. The simulations are performed on a number of meshes of different density using a number of turbulence models. It is found that k-epsilon two-equation turbulence model with a second-order algorithm on a reasonable mesh gives the best results. This information is then used to determine the modeling requirements (mesh, numerical algorithm, turbulence model etc.) for flows in 3D enclosures with different ventilation systems. In particular two cases are considered for which the experimental data is available. These cases are (1) air flow and heat transfer in a naturally ventilated room and (2) airflow and temperature distribution in an atrium. Good agreement with the experimental data and computations of other investigators is obtained.

Sign in / Sign up

Export Citation Format

Share Document