renewable energy

4 downloads 0 Views 1MB Size Report
prov1s10~. Th~ forthcoming rest~iction on the use of CFC gases for r,:frl~eratlon and the product10n of CO , which is a major gas ..... r¢. is the difusion coeficient.
RENEWABLE ENERGY TECHNOLOGY AND THE ENVIRONMENT

VOLUME 4 Solar and Low Energy Architecture

Edited by

A. A. M. SAYIGH

PERGAMON PRESS

(ii) The control of the air flow is environmental c (iii) Because of the consumption can ventilation sys more difficult t

SIMULATION OF SOLAR-INDUCED VENTILATION

(iv) Supply of clean polluted cities.

H B Awbi and G Gan Department of Construction Management & Engineering university of Reading, UK

ABSTRACT Computational fluid dynamics (erD) is used to simulate the air flow and heat transfer in a Trombe wall and a solar chimney. The crn results are compared with calculations using heat transfer and fluid flow equations for a closed channel with good agreement.

KEYWORDS Passive ventilation, computational fluid dynamics, simulation, solar air collectors, solar chimneys.

computer

INTRODUCTION The information technology revolution of the 1980 I S and the tendency to build deep-plan office buildings have attributed to large increases in internal heat production in commercial buildings. As a result, there has been a vast expansion in the use of airconditioning systems to control the internal environment of such buildings. Although the expansion in the use of x:efrigeration ,systems :to cool buildings is expected to contlnue for somet1me yet, 1n the long term bUildings will have to r~l¥ more on passive ventilation for cooling and fresh air prov1s10~. Th~ forthcoming rest~iction on the use of CFC gases for r,:frl~eratlon and the product10n of CO , which is a major gas contr1butlng to,the greenhouse effect, wil~ inspire designers to apply more passlve means for controlling the indoor environment. Passi ve ventilation for commercial buildings has not been popular in the past mainly due to the following reasons: (i) Outdoor air is usually only accessible to perimeter zones of about 6rn from windows or ventilation openings.

2016

Most of these problems, r( more intelligent designs , are now available at low c( passively ventilated using controlling the air flow i Most natural ventilation sy to create the necessary air in the absence of wind, bue provide the required air building fabric using sola enhance the influence of bu between fabric and air inc I aI, 1977) and the solar chim shown to be effective methc these systems use vertica effective at lower solar a loc~tions in most seasons, reglons for most of the yea The performance of the Troml studied over long periods t Sweeney and Wormald, 1988) the heat transfer charac1 Akberzadeh et al. 1982). Sj the absorber wall have also solution of the conductio thickness for the storage an Wormald, 1988). The flow an been simulated (Borgers and A and energy equations using mixing length turbulence mod In this paper, a computati developed by the authors is heat transfer in a Trombe required for ventilation in ... in summer. The flow in a 5 summer ventilation. The calculations using heat trans channel. Considering the s' empirical solutions the ag generally good.

(ii) The control of ventilation rates is difficult because the air flow is largely influenced by external environmental conditions. (iii) Because of the lack of control, the energy consumption can be greater than for mechanical ventilation systems and thermal comfort could also be more difficult to achieve.

lCED VENTILATION

(iv) Supply of clean outdoor air may be difficult in airpolluted cities.

G Gan lagement & Engineering ding, UK

s used to simulate the air 11 and a solar chimney. The at ions using heat transfer ~losed channel with good

fluid dynamics, ar chimneys.

computer

,n of the 1980's and the Lldings have attributed to 9roduction in commercial ~n a vast expansion in the ) control the internal rh the expansion in the use uildings is expected to term buildings will have 'or cooling and fresh air n on the use of CFC gases , CO , Which is a major gas will inspire designers to :g the indoor environment. :Ungs has not been popular :; reasons: ~ccessible

to perimeter or ventilation openings.

Most of these problems, real as they are, could be overcome by more intelligent designs and by advanced control systems which are now available at low cost. Even some ancient buildings were passively ventilated using ingenious methods of harnessing and controlling the air flow into the building (Bahadori, 1979). Most natural ventilation systems rely on wind and buoyancy forces to create the necessary air motion through the building. However, in the absence of wind, buoyancy alone may not be sufficient to provide the required air flow rates. However, by heating the building fabric using solar irradiance it will be possible to enhance the influence of buoyancy as the temperature difference between fabric and air increases. The Trombe wall (Trombe, et aI, 1977) and the solar chimney (Bollchair, et aI, 1988) have been shown to be effective methods of increasing buoyancy. Both of these systems use vertical solar collectors which are more effective at lower solar altitudes which would occur in arid locations in most seasons, except the summer, and in temperate regions for most of the year. The performance of the Trombe wall as an air collector has been stUdied over long periods to predict overall performance (e.g. Sweeney and Wormald, 1988) and over short periods to determine the heat transfer characteristics of the collector (e.g. Akberzadeh et al. 1982). Simulations of the heat conduction in the absorber wall have also been carried out using a numerical solution of the conduction equation to optimize the wall thickness for the storage and emission of heat (e.g. Sweeney and Wormald, 1988). The flow and heat convection in the air gap has been simUlated (Borgers and Akbari, 1984) by solving the momentum and energy equations using a finite difference method and the mixing length turbulence model. In this paper, a computational fluid dynamics (CFD) program developed by the authors is used to simulate the air flow and heat transfer in a Trombe wall for heating the outside air required for ventilation in winter and for inducing ventilation in summer. The flow in a solar chimney is also simulated for summer ventilation. The CPD results are compared with calCUlations using heat transfer and flow equations for a closed channel. Considering the simplifying assumptions made in the empirical solutions the agreement with the CFO results is generally good.

2017

where W is the channel two walls. For a narrow

AIR COLLECTOR PERFORMANCE The performance of an air collector for the ventilation of a building is assessed by the air flow rate that can be provided, the air supply temperature to the building and the diurnal and seasonal variations of these quantities. 'l'hese performance parameters can be evaluated using either simple heat transfer and air flow expressions and empirical data or a full CFD simulation of the flow. Both of these techniques are considered here and the results obtained from them are compared in the next section.

The convective heat air is given by: Q = where he is the surface area and T w and air. The heat transfer Nu

simplified Analytical Approach Considering buoyancy effects on It (i.e. excluding wind effects) the volume of air flow rate, V (m /s), produced by a vertical air collector is given by:

for turbulent flow (1013 Nu for laminar flow (10 9 > Rayleigh number Ra are

( 1)

are the inlet and mean air temperatures of the collect~r (K) / mH is the height between inlet and outlet openings and (Cd A) is the product of effective discharge coefficient and area of the collector openings, Fig. 5(a). This is given by:

Nu Ra

where T. and T

n (CdA),=E

(2 )

(Cd A),

i=l

where Pr

MCp/k =

Gr

gB y 3 (T w

-

T ai )

The air temperature in inlet is obtained from:

for openings in parallel, and where A 1 (3 )

B

hw + hg

and subscripts wand 9 wall), T!I\ i~ the inlet rate of alr ln the only, equation (9)

for openings in series, where n is the number of openings. The pressure losses in the collector can be evaluated using: H

Am

Am

bP = [ 4f - - + K, ( - - ) + Kd ( - ) + K, Dh A, Ad

Am (--)

2

J ~PmVm2

(4 )

A,

CFD SOLUTION

where the K's are the pressure loss coefficients, the A's are the flow areas, H is the height between the top and bottom openings, p is the air density and V is the mean velocity. SUbscripts i, d, 0 and m refer to inlet, damper (or other fittings), outlet and channel respectively. The hydraulic diameter of the channel, 0h' is given by: 2 \1 h (5) \V

+ h

2018

Further details are

The analysis given earl air collectors but does patterns in the latter requires the (CPO) program. Here / a cFO program predicting the air flow space. The program is which uses the k-~

where W is the channel width and h is the distance between the two walls. For a narrow channel (i.e. W » h) Dh Z 2h. the ventilation of a that can be provided, g and the diurnal and These performance rnple heat transfer and a full CFD simulation e considered here and j in the next section.

The convective heat transfer between the collector surface and air is given by: (6)

where he is the convective heat transfer coefficient, A is the surface area and Tw and Tai are the temperatures of wall and inlet air. The heat transfer coefficient he can be evaluated using: Nu = 0.09 Ra 1/3

:cluding wind effects) uced by a vertical air

(7)

for turbulent flow (10 13 > Ra > 10 9 ), and Nu = O. 53 Ra 1/4

(8)

for laminar flow (10 9 > Ra > 10 4). The Nusselt number Nu and the Rayleigh number Ra are given by:

(1) r temperatures of the and outlet openings :harge coefficient and This is given by: ~t

(2)

Nu Ra where Pr

Gr

Pr Gr

llCplk = Prandtl l s number

gBy 3(Tw

-

Tii)/U Z = Grashof's number

The air temperature in the collector at any height y above the inlet is obtained from: T, ~ A/B +

(T"

-

A/B)

exp(-B W

y/rnc p )

(9)

where A ( 3)

and sUbscripts wand g refer to wall and glazing (as in a Trombe wall), Tal is the inlet air temperature and rn is the mass flow rate of a1r in the collector. For heat transfer from one surface only, equation (9) reduces to:

lber of openings. )e evaluated using:

A. (--)

2 ]

%Pmv..2

B

Ta = T w + (T. i

(4)

A.

-

Tw)

exp(-h w W y/mc p )

(10)

Further details are given in Appendices (A) and (B). CFD SOLUTION

ients, the Als are the ? and bottom openings, ocity. Subscripts i, fittings), outlet and .er of the channel, Dh ,

The analysis given earlier can be used for the slZlng of vertical air collectors but does not provide information on the air flow patterns in the collector or the building which it serves. The latter requires the application of a computational fluid dynamics (CFD) program.

(5)

Here, a CFO program developed by the authors is used for predicting the air flow within the collector and the ventilated space. The program is a three-dimensional finite volume code which uses the k-e turbulence model and solves the Navier-Stokes

2019

equations t the energy equation and the equations for the turbulence energy (k) and its dissipation (€). These equations have the general form:

8 -(pU,1')

aX

j

convection

8

81'

-(r.--)

aX

j

aX i

diffusion

+ S,

Table 1 Calculated and sj

(11)

source

where the dependent variable ¢ represents the velocity components U, V and W, the temperature T, the turbulence scales k and € and r¢. is the difusion coeficient. The source term S is given by different expressions for each ¢. A staggered gri~ is used for solving the equations using the SIMPLE pressure correction procedure. At. solid. boun~aries wall function expressions are used for the f1rst gr1d p01nt from the boundary. The iterative solution procedure is repeated until a converged solution is achieved (Awbi, 1991, Awbi and Gan, 1991). ~he~

a CFD program is used for predicting buoyancy-driven flows 1t 15 necessary to know the pressure difference between the air inlet and outlet openings which is required for Bernoulli I 5 equa~ion to specify the air flow rate into the system. Equation (1) 1S used to calculate the flow into the collector for every iteration. RESULTS AND DISCUSSION Trombe Wall Trombe wall collectors have traditionally been used for space heating by allowing air from the room to enter near the bottom of the wall which is heated by the collector and then returned back to the room at high level. This does not normally include fresh air supply to the room and assumes that fresh air is prov~ded fortuitously by air infiltration through the building f

-G- Two heoled wans ... Onehcawdw:llJ

02

0.4

0.6

Buoyancy Equation

The buoyancy (stack)

pres~

datum is: p

='

Po

,,+--~-~-~~-~---{

00 0.0

-G- Two healed walls ... Onch""llllI wall

!:: 1-~~:::::::::P . . .-o=..d~

.,g 02 0.'

40

0.'

0.0

0.2

0.4

,jW

0.6

0.8

1.0

where p (kgjm 3 )

,jW

Vdocll}' promo

Temperature prom"

.0

is the datum pre: The

pressure

separated by a vertical

Fig. 7 Velocity and temperature profiles at the exit from u ~'olar chlnmcy

d~

dp = - ppg

where p is the air densit~ T is th~ air temperature !

0.10

,.:t S >

0.00

~

0.'"

,, ,

~



0.08

Integrating Equation (A.2 between the lower and upp 5 (a) :

0,02

Mo"-,,,rcd by Bouch.;, c\ al.



Simolaloo

.00

where H is the vertical ~ and Tc are the air temper9

20 Tw(clcg.C)

Fig. 6 Effect of Chimney wall temperature on the flow ralc

The air volume flow rate,"

v=

(Cd A)

i

(Cd A) c is the prod\ and area of the flow open~

where

(A.3) and taking the refel we find that:

V=

Fig. 8

Predicted velocity vectors and

isotherms in a solar chimney Left: two heated walls Right: one heated wall

Temperature: inlet air heated wall unheated wall i.