Original articles
Lattice Boltzmann simulation of natural convection in a square enclosure with discrete heating

https://doi.org/10.1016/j.matcom.2020.07.025Get rights and content

Abstract

In the present work, natural convection heat transfer in a differentially heated cavity is considered, in which, source–sink pairs are located on vertical walls with a constant temperature while other wall parts of the enclosure are insulated. The effects of different arrangements, sizes, and number of heat source–sink pairs on the flow and thermal fields are investigated by using the Lattice Boltzmann Method to solve the flow and thermal field equations. In this paper, streamlines, isotherms, the average Nusselt number, and the entropy generation are presented to get a better insight into the nature of the problem. Results of this study showed that the arrangement of the heat source–sink pairs could strongly affect the flow and thermal fields in the cavity due to the formation of the vortices. It was also found that the highest amount of heat transfer with the lowest entropy generation was achieved by splitting the discrete heat source–sink pairs into the smaller segments and putting them alternately on one sidewall.

Introduction

Natural convection can be seen in a wide range of applications such as an electronic component cooling system since it can be regarded as a cheap, safe and noise-free mechanism. In this mechanism, a temperature gradient in a flow field causes density differences which can induce the natural convection heat transfer. Many research projects have been conducted to optimize this heat transfer mechanism [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11]. Among these studies, a rectangular enclosure with discrete heat source–sink pairs seems to be an interesting topic among scientists because it has many industrial applications in different fields. In these rectangular enclosures, natural convection is proved dominant heat transfer mechanism and flow field structure is strongly affected by the arrangements and locations of the discrete heat source–sink pairs. Some of the studies in this field are as follows.

Sezai and Mohamad [9] studied the effects of different length and width ratios of heat sources in a horizontal enclosure. Their results showed that the maximum and minimum heat transfer rates were achieved at the edges of the discrete heat sources and at the center of the enclosure, respectively. Mezrhab et al. [5] conducted research on the inclined differentially heated square cavity by using numerical simulation. Their results indicated that inclined cavities had higher average Nu numbers than vertical ones for a low Rayleigh (Ra) numbers, while opposite trends were seen for high Ra numbers. Abu-Nada et al. [1] studied the effects of inclination angle and volume fraction of nanoparticle in a square enclosure. They found that the addition of nanoparticles remarkably enhanced the heat transfer in comparison with pure fluid. This trend was more significant for low Ra numbers. Furthermore, their results proved that heat transfer could be controlled by an inclination angle. Mirabedin et al. [6] investigated the effects of the Ra number and a central angle on natural convection heat transfer in circular enclosures. They found that the heat transfer enhanced by increasing the Ra number and decreasing the central angle. It was also seen that there was a linear trend in heat transfer rate after achieving critical Ra numbers. Baghsaz et al. [2] conducted research on natural convection heat transfer in a porous cavity filled with nanofluid by considering entropy generation. Their results indicated that irreversibility of the process and natural convection heat transfer increased by increasing the values of porosity and Ra numbers. Zhan et al. [11] studied the natural convective heat transfer in a closed cavity by using an experimental approach and reported that, by increasing the distance between heat sources, the heat transfer was improved and then decreased. Their studies also proved that heat transfer could be enhanced by increasing the temperature differences between heat sources and sinks.

Reviewing the previous literature has shown that no attempts have been made in order to consider the effects of different number and arrangements of discrete heat source–sink pairs on the natural convection heat transfer and entropy generation in a square cavity using the Lattice Boltzmann Method (LBM). Therefore, in the present work, two and three heat source–sink pairs at different Ra numbers are considered to find the best case.

Section snippets

Mathematical formulation

In this paper, the problem of interest is to simulate natural convection heat transfer in a two-dimensional square enclosure with discrete heating on the vertical sidewalls and filled with air. The flow inside the cavity is considered steady and laminar (Ra=102107). The governing equations involve the continuity, conservation of momentum, and energy equations.

Numerical method and its validation

In this study, the natural convection heat transfer in a square cavity is studied using LBM implemented by an in-house computer code written in FORTRAN. In order to consider the mesh independence study and validation of the model, the results of the work done by Davis [10] are selected. In [10] natural convection in a square cavity is studied where Pr=0.707. In addition, the temperatures of the vertical walls are assumed to be Th=1 and Tc=0 while the horizontal walls are insulated (Fig. 1).

Hydrodynamic and thermal analysis

As mentioned above and illustrated in Fig. 2, two and three heat source–sink pairs with three different arrangements for each of them are studied in this paper by considering their effects on the flow field, temperature field and rate of heat transfer. In this work, the Pr number is kept constant at 0.707 and the Ra number is varied from 103 to 107.

Streamlines and isotherms for a given Ra number are depicted in Fig. 3, Fig. 4 for different cases. These figures show that the flow moves upwards

Conclusions

Natural convection heat transfer in a square enclosure with discrete heat source–sink pairs is simulated in this work by using LBM. The effects of the different arrangements, and number of heat source–sink pairs and Ra numbers are considered and listed as follows:

  • The flow is not affected much by thermal boundary layers in the central region of the cavity and the temperature gradient is not steep.

  • Temperature gradients are severe in the vicinity of the heat source–sink pairs due to the contracted

Nomenclature

cLattice speed, m/s
csSpeed of sound, m/s
fDensity distribution function
feqEquilibrium density distribution function
gEnergy distribution function
geqEquilibrium energy distribution function
gGravitational acceleration
HHeight of enclosure
NuNusselt number
PPressure
PrPrandtl number
RaRayleigh number
SVolumetric rate of entropy generation, W(m3K)
SDimensionless volumetric entropy generation rate
TTemperature
UDimensionless velocity component in X direction
VDimensionless velocity component in Y direction
X

Cited by (20)

  • Exponential nonuniform wall heating of a square cavity and natural convection

    2022, Chinese Journal of Physics
    Citation Excerpt :

    Apart from these, we will be rather concentrating on the recent progress. The main topical contributions to the convection phenomena are the impacts of solid particle dispersions [11], the uniformly versus non uniformly heated walls [12], the partially active side walls [13], the tilted square cavities [14], the partial heating and cooling at vertical walls [15], the non-uniform wall heating with CuO nanofluid [16], the concave/convex side walls [17], the ferrofluid analysis [18], the effect of an obstacle [19], the non-uniform heating [20] the magnetic field effects [21], the c-shaped enclosure saturated by a nanofluid [22], the flexible side wall impacts [23], the influence of electric field [24], the u-shaped cavity [25], the differentially heated enclosures with variable temperatures [26], the multiple heated blocks [27], the effects of a complicated fin [28], the scaling analysis of natural convection in a vertical channel [29], the influences of internal heating, strong turbulence and phase change [30], the horizontal annular elliptical finned tubes [31], the open-ended vertical parallel [32], the H-shaped cavity with two-sided inclined porous fins [33], the hybrid nanofluid [34], the thermophoresis and Brownian motion effects [35], the low porosity cylindrical annuli [36], the presence of a rotating cylinder [37], the non-isothermal walls [38], the magnetic field from a single permanent magnet [39], the discretely heated inner body [40], the wall superheat [41], the thermogravitational convection [42] and the Lattice Boltzmann simulation [43]. Although a variety of past research was numerically completed on the natural convection with uniform wall heating some of which are aforementioned, less attention was paid on the nonuniform wall heating case; such as the sawtooth temperature distributions [44], the divided higher and lower temperature [45], the sinusoidally applied temperature [46–52], and the linear heating [53,54].

  • Thermal and entropy analyses on buoyancy-driven flow of nanofluid inside a porous enclosure with two square cylinders: Finite element method

    2021, Case Studies in Thermal Engineering
    Citation Excerpt :

    Further, a remarkable increase in current is obtained for high Rayleigh numbers regardless with the blade length and its position. The impact of diverse sizes, arrays, and number of heat sink-source couplings on the NC within a container loaded with air was explored by Dadvand et al. [6]. In an experimental work, Wu and Ching [7] presented their work through a cavity filled with air and there is a restriction on the top of the wall.

  • Influence of dome shape on flow structure, natural convection and entropy generation in enclosures at different inclinations: A comparative study

    2021, International Journal of Mechanical Sciences
    Citation Excerpt :

    Additionally, it is considered in various industrial sectors such as aerospace industry, metallurgical and petrochemical processes [6]. With the rapid development of technology and restrictions on energy utilization, the appeal of natural convection is rapidly growing as it is a safe, low-cost and silent heat transfer mechanism that requires no external energy, and it is induced by buoyancy forces resulting from the density variations due to the temperature differences in a fluid medium [7–9]. Considering the importance of natural convective heat transfer in enclosures for engineering applications, researchers have carried out numerous experimental and numerical studies to achieve significant progress for the improvement of natural convective heat transfer performance [10,11].

View all citing articles on Scopus
View full text