science direct traducido al spanish

Journal of Building Engineering 4 (2015) 52–59

Contents lists available at ScienceDirect

Journal of Building Engineering

journal homepage: www.elsevier.com/locate/jobe

Investigation of laminar natural convection heat transfer within tubular daylighting devices for winter conditions

Tolga Pirasaci n

Department of Mechanical Engineering, Gazi University, Ankara, Turkey

a r t i c l e i n f o

Article history:Received 8 April 2015Received in revised form28 July 2015Accepted 6 August 2015Available online 10 August 2015

Keywords: Daylight Light-pipesTubular Daylighting DevicesNatural convection

a b s t r a c t

Recent developments in lighting and energy efficiency such as Tubular Daylighting Devices (TDD) aim at reducing energy consumption and providing homogeneous illumination in buildings. This ensures energy savings by reducing lighting energy consumption.

To prevent the increase of total energy consumption, the heat loss at the TDD should be taken into consideration when using TDD. This paper presents an experimental and numerical study on the laminar natural convection in TDD for winter conditions. The results show that the overall heat transfer coefficient of TDD can be decreased by using a separator plate in the TDD. Moreover, the overall heat transfer coefficient changes significantly with the position of the separator plate.

& 2015 Elsevier Ltd. All rights reserved.

1. Introduction

Today's life style forces some people to live in places in- sufficiently illuminated. Experimental studies show that insufficient daylight may result in psychological and physiological problems in humans while adversely affecting work efficiency [1–3].

Recent developments in lighting and energy efficiency such asoptical daylighting systems aim at reducing energy consumption and providing homogeneous illumination in buildings. One of these strategies is the use of daylight transmission systems with high efficiency and low maintenance costs. By using these systems sufficient lighting can be provided and electricity consumption of lighting is reduced. Daylight transmission systems are used to homogenize the illumination level, increase visual comfort and achieve energy savings of the space [4–8].

One of these systems is the Tubular Daylighting Device (TDD). TDD transmits the sunlight from skylight to the space by using a reflective channel giving satisfactory results for the areas of the buildings where the sunlight cannot reach [9–11].

TDD (Fig. 1) are composed of five main parts; dome, dome base,roof base, reflective channel and diffuser. In these systems the acrylic dome is placed on the roof and transmits the sunlight to the reflective channel. The sunlight coming into the channel is reflected to the diffuser which provides natural lighting by dis- tributing the sunlight homogeneously.

n Fax: þ 90 312 2319810.E-mail address: [email protected]: http://w3.gazi.edu.tr/ pirasaci/

In the design on these systems the heat loss at the TDD should be taken into consideration. Otherwise, the amount of energy required for heating may be greater than the lighting energy saved and thus the building's total energy consumption may increase.

Although some overall heat transfer coefficient values are givenin product catalogs, no papers were found about the heat losses in the TDD. As well as there being very few studies on the heat transfer occurring in the dome skylight which is the closest in geometry to the TDD in the literature.

One of these studies was conducted by McGowan et al. [12]They investigated the thermal performance of pyramidal and barrel vault skylights by conducting measurements and numerical simulations using a commercial computational fluid dynamics (CFD) package. Another study was presented by Klems [13] in which nighttime measurements of the net heat flow through several types of skylights were presented and the measured U- values were compared with calculations using WINDOW4 and THERM programs. In the studies presented by Laouadi et al. [14,15] the laminar natural convection within concentric domed cavities was investigated by using the numerical control volume approach. Natural convection heat transfer in horizontal fully hemispheric domed cavities with planar inner surfaces was studied by Saber and Laouadi [16]. Their numerical model was based on the finite- element method. Saber et al. [17] recently studied using convective heat transfer in low-profile spherical cavities with planar bottom surfaces by using a finite-element method.

Literature survey shows that there is limited information aboutthe heat transfer in TDDs. In the present research the natural convection heat transfer occurring TDDs was examined for winter

http://dx.doi.org/10.1016/j.jobe.2015.08.0032352-7102/& 2015 Elsevier Ltd. All rights reserved.

T. Pirasaci / Journal of Building Engineering 4 (2015) 55 T. Pirasaci / Journal of Building Engineering 4 (2015)

Nomenclature wall 4, W

A surface area, m2

Q B experimental conduction heat transfer rate from base, W

ATDD cross sectional area of the TDD, m2 Q S.P. experimental conduction heat transfer rate from sur-cp

Dspecific heat, kJ/kgKdiameter of the TDD, m T

round panel, Wtemperature, K

k thermal conductivity, W/mK Tm.c. metering chamber temperature, KL thickness, m Tc.c . climatic chamber temperature, KP pressure, Pa Tin average temperature of the inner side of insulation, KQ TDD total heat transfer rate, it W Tout average temperature of the outer side of insulation, KQ H measured electrical power supplied to the heaters, W UTDD overall heat transfer coefficient, W/m2KQ S.1 experimental conduction heat transfer rate from side

wall 1, WQ S.2 experimental conduction heat transfer rate from side



Greek symbols

β thermal expansion coefficient, 1/Kν kinematic viscosity, m2/sρ density, kg/m3

Dome

Dome Base

Roof Base

ReflectiveChannel

Diffuser

Fig. 1. Tubular daylighting device.

conditions. For this purpose, the test system was established and the overall heat transfer coefficient was determined by testing TDD. Then, numerical studies were performed, and thermal transmittance coefficients of various TDD configurations were calculated.

2. Experimental set-up and data reduction

Fig. 2 shows a schematic representation of the experimental set-up which is composed of climatic chamber with refrigeration unit, metering chamber, surround panel, heaters, controller and

measurement systems.The climatic chamber is an open base cabinet, and the re-

frigeration unit placed at the top of this cabinet. This chamber is used for simulating winter outdoor conditions. For this purpose indoor temperature of the chamber was stabilized at 18 °C with a refrigeration unit during the experiments. The dimensions of the climatic chamber are 980 mm 980 mm 650 mm [inner dimensions (width depth length) ]. All chamber walls are constructed from 10 mm Plywood þ 50 mm Styrofoam þ 10 mm Ply- wood plates.

The metering chamber is an open ceiling cabinet and used for simulating winter indoor conditions. The dimensions of this chamber are 980 mm 980 mm 650 mm [inner dimensions (width depth length)]. All chamber walls are constructed from10 mm Plywood þ 50 mm Styrofoam þ 10 mm Plywood plates. During the experiments chamber temperature was stabilized at21 °C with a heating system. Heating system consisted of 2 heaters, 1 PID temperature controller and a watt-meter. Watt-meter was used for the measurement of the supplied electric power to these heaters.

The surround panel is mounted and placed between climatic and metering chambers. The dimensions of the surround panel are1100 mm 1100 mm 270 mm (width depth length) and constructed from 10 mm Plywood þ 5 50 mm Styrofoam þ 10 mm Plywood plates.

Ninety thermocouples were used for temperature measurements. All thermocouples were separately calibrated. Signals from the thermocouples were collected, processed and stored with computer connected seven ELIMKO 680 series universal data loggers. Temperature readings were taken at several locations on the bottom (8 inner side and 8 outer side of styrofoam insulation) and side (8 inner side and 8 outer side of styrofoam insulation) walls of the metering chamber and at several locations on the surround panel (4 inner side and 4 outer side of styrofoam insulation). Indoor temperatures of chambers and ambient temperature were also measured.

The experiments were carried out when the ambient temperature is below 21 °C. Initially all setup was at the thermal equilibrium with the ambient air. After the heaters and refrigeration unit were turned on, the temperature of metering chamber increases and the climatic chamber temperature decreases. The increase in temperature of the metering chamber continues until it reaches 21 °C. At this temperature it was stabilized by using a controller. Similarly the temperature of the climatic chamber was


Refrigeration Unit

-18oCPlywood

Styrofoam

Tubular DaylightingDevice (TDD)

Thermocouples

Surround Panel

Computer

Metering Chamber21oC

Heater

PID Temp. Controller Watt-meter

~Fig. 2. Experimental setup.

stabilized at 18 °C during the experiments. The experiments were continued until steady-state conditions. It was observed that experimental conditions reach a steady-state condition after ap- proximately 4–6 h. After conditions had been steady for some timeand differences in temperatures between two intervals became

uncertainty analysis was conducted on all measured quantities as well as the quantities calculated from the measurement results. Uncertainties were estimated according to the standard proce- dures reported in the literature see e.g. [18–22]. Overall, the uncertainty in the overall heat transfer coefficient UTDD is around

negligible (ΔT < 0.1 °C), all temperatures were collected and

7 6%.

stored. Using this data, heat transfer calculations were done as follows.

The overall heat transfer coefficient UTDD was calculated fromthe following equation:

U = Q T D D

3. Numerical formulation and solution procedure

The natural convective flow in the TDD is assumed to be two- dimensional, axisymmetric (Fig. 3), laminar, incompressible, stea-TDD

A (T − T ) (1) dy and all thermo physical properties are constant, except for fluidTDD m . c . c . c .

fluid density is treated by using Boussinesq approx-where Q TDD is the total heat transfer rate, ATDD is the cross sectional area ( πD2/4) of the TDD, D is the diameter of the TDD and Tm .c. − Tc.c. is the difference between metering and climatic chamber temperatures.

density. Theimation. In addition the radiation heat transfer between surfaces is not accounted for in this study. This approach has been success- fully applied to natural flows in enclosed cavities [16,17]. Following

fluid areThe total heat transfer rate (Q TDD ) from TDD was calculated these assumptions, the conservation equations for the

from an energy balance written for the metering chamber and expressed as given below:

∂ur

∂r+ ∂u z

∂z= 0 (4)

Q TDD = Q H. − Q S.1 − Q S.2 − Q S.3 − Q S.4 − Q B. − Q S . P.

(2) u u u u 1 P 2u 2u∂( r r ) + ∂( z r ) = − ∂ + ν ( ∂ r + ∂ r )

where Q H is the measured electrical power supplied to the heaters, Q S.1, Q S.2, Q S.3, Q S.4 , Q B. and Q S.P. are the conduction heattransfer rates from side walls, base wall and surround panel

∂r

∂ (ur u z )

∂z

∂ (u z u z )

ρ ∂r

1 ∂P

∂r2

∂ 2u z

∂z2

∂ 2u z

(5)

respectively.These conduction heat transfer rates were calculated from

+∂r ∂z= − + ν (

ρ ∂z ∂r2 + ∂z2

) + gβ (T − T0 ) (6) Fourier's law (Eq. (3)) according to the measured surface tem- ∂ (ur T) + ∂ (u z T) k

∂ 2T∂ 2T

L

T. Pirasaci / Journal of Building Engineering 4 (2015) 55 T. Pirasaci / Journal of Building Engineering 4 (2015) peratures and known material properties: ∂r ∂z ρcp ∂r2 ∂z2 (7) k i A iQ = (T − T ) The conjugate steady conduction in the solid separator platei in out i

i (3) body is coupled to the natural convection. In these regions the

where the subscript i is the surface index ( S.1, S.2, S.3, S.4, B. and conservation of energy is defined asS. P ), k is the insulation thermal conductivity (for Styrofoam in- ∂ 2T ∂ 2Tsulation 0.035 W/mK), A is the surface area, L is the insulation thickness (0.05 m for sides and bottom, 0.25 m for surround panel

∂r2 +

∂z2 = 0 (8)

), Tin is the average temperature of the inner side of insulation andTout is the average temperature of the outer side of insulation.

In order to determine the reliability of experimental results, an

No slip and the continuity of normal heat flux boundary con-ditions are applied at the fluid/solid interfaces while no slip and constant temperature boundary conditions applied at the outer

c


dur = uz= 0 T = Toutdoor

second order upwind scheme was used for discretization of energy and momentum equations and PRESTO algorithm was used for the pressure correction. The velocity–pressure coupling was established by using SIMPLE algorithm.

To get converged solution, iterations were continued until all residuals were less than 1 × 10−6 and the total heat flux of all borders were less than 1 × 10−3. Grid independence tests showthat for 13,139 and larger mesh numbers the numerical results did not change anymore. Thus the analyses were performed for the grid structure (Fig. 4) having 13,139 mesh number.

aur = 0

∂ T ∂r

=0

z

b ur = uz= 0 qcond. = qconv.

Tsolid = Tfluid

r

ur = 0 ∂T ∂r

=0

dur = uz= 0 T = Tindoor

4. Results and discussion

4.1. Experimental results

Natural convection heat transfer in a TDD was investigated experimentally for two configurations of TDD. In the first configuration the standard TDD having 550 mm tube diameter (Fig. 5a ) was tested between 21 °C and 18 °C. In the second configuration the acrylic separator plate (10 mm thick) was placed at the top of the channel for separating the natural convection flow in the TDD. The schematic illustration of this configuration is shown in Fig. 5b.

As a consequence of the above mentioned experimental conditions, the overall heat transfer coefficients, UTDD, were obtained as 3.78 W/m2 K and 2.34 W/m2 K for configurations 1 and 2,respectively.

Experimental results show that the heat transfer from TDD canFig. 3. Schematic representation of numerical model with boundary conditions(a) symmetry (b) fluid/solid interface (c) insulated wall (d) wall at constant temperature.

Fig. 4. CFD meshing model.

surfaces. All applied boundary conditions are shown in detail atFig. 3.

Here ur and uz are the velocities at the r and z directions respectively, q is the heat flux and T is the temperature.

In this study numerical analysis was performed by using a commercial CFD package FLUENT. All conservation equations (Eqs. (4)–(8)) were discretized using the control volume approach. The

be decreased by using a separator plate.

4.2. Numerical results

In this study the natural convection heat transfer in the TDD was analyzed numerically and the impact of the usage of separator plate on the natural convection flow and heat transfer was examined.

At the beginning of the numerical study analyses were performed for both of the TDD configurations 1 and 2 which were investigated experimentally before. The study continued with the cases in which the placement of separator plate was examined. In Fig. 6 the placement of separator plate is shown schematically for different configurations.

In all configurations the outer (dome and base zone) and the inner (channel and diffuser zone) walls were kept at the constant temperature of 18 °C and 21 °C, respectively. Under these conditions, the flow was stable and reached the steady state. After all analyses were performed heat transfer rates from TDD and the overall heat transfer coefficient UTDD were calculated and the op- timum placement of separator plate was determined.

For the numerical model validation, the results obtained using the numerical model were compared with the present experimental results for TDD configurations 1 and 2 (Table 1). Compar- ison of numerical and experimental UTDD values are listed in Table 1.

As shown in Table 1 the numerical results agree very well with the experimental results with an error of 2%. Moreover the overall heat transfer coefficient in configuration 2 is 62% of that in configuration 1.

The comparison of the numerical results for configurations1 and 2 is shown in Fig. 7.

In Fig. 7a and d the direction and the ratio of heat transfer is given. It is seen from these figures that the heat is transferred from indoor environment to the air in TDD over the channel and diffuser zones then it is transferred to the outdoor environment over the dome and base zones. The ratios given here are the heat transfer rates from zone walls (Q) to the total heat transfer rate


a

DomeZone

Configuration 2 ----- Configuration 3 ----- Configuration 4 -----

Configuration 5 -----







Configuration 12 -----z

Configuration 13 ----b

r

BaseZone

InsulatedZone

ChannelZone

DiffuserZone

Fig. 6. Placement of the separator plate in the TDD for different configurations.

Table 1Comparison of numerical and experimental UTDD values.

Experimental (W/m2 K) Numerical (W/m2 K) Error (%)

Configuration 1 3.78Configuration 2 2.34

transfer from indoor environment. This causes the increase in buoyancy force and results in the air flow in the upward direction near the TDD wall. Meanwhile very low temperatures at the dome wall cause secondary flow directed to the downward near the dome wall. The mixture of these streams results in a relatively coldstream having the temperature of ∼275 K. This stream moves

Fig. 5. Schematic illustration of tested TDD. (a) Configuration 1( b) Configuration2.

from configuration 1 (Q0). Total heat transfer can be calculated by adding heat transfer rates of the channel and diffuser zones. This results show that in both configurations the heat transferred from channel zone is greater than diffuser zone as expected. Because the diffuser is made of acrylic (k ¼ 0.2) whereas the channel is made of aluminum (k ¼ 237) and the area of the channel zone is greater than the diffuser zone. For outer (dome and base) zones however, there is a difference between two configurations. In configuration 1 the heat transferred from dome zone is greater than base zone. But in configuration 2 the heat transferred from base zone is greater than dome zone.

This difference can be explained by considering velocity and temperature distributions (Fig. 7b, c, e, f) in TDD. It is seen from these figures that in configuration 1 the air temperature gets higher values near the channel and diffuser zones due to the heat

downward at the center of the TDD. This flow pattern causes low temperatures ( ∼275 K) and higher heat fluxes in the TDD. In configuration 2, however, separator plate divides TDD into two volumes. In lower volume the air temperature gets higher valuesnear the channel and diffuser zones due to the heat transfer from indoor environment. This causes an increase in buoyancy force which produces the air flow in the upward direction near the TDD wall. When the flow reaches base zone wall and separator plate, it losses heat and its temperature drops. The cold air moves downward at the center of the TDD. The heat conducted through the separator plate transferred to the outdoor environment by the flow in the upper volume. Due to the low conductivity of acrylic separator plate and these flow patterns the average temperature inconfiguration 2 ( ∼286 K) is greater than the average temperaturein configuration 1.

The comparison of overall heat transfer coefficients (UTDD) for different configurations is given in Fig. 8.

It is seen from the figure that the lowest overall heat transfer coefficients (UTDD) obtained for the configurations in which the separator plate is placed in insulated zone and the minimum value of these coefficients is for configuration 6. In configuration 6 the

Con

figur

atio

n C

onfig

urat

ion


a b c57%

DomeZone

BaseZone

43%

InsulatedZone

ChannelZone

Vel. (m/s) Temp. (K)

z r

37%

63%

DiffuserZone

d e f24%

DomeZone

BaseZone

38%

InsulatedZone

ChannelZone

z r

23%

39%

DiffuserZone

Fig. 7. Numerical results for configurations 1 and 2.

separator plate is placed at the top of the insulated zone. The overall heat transfer coefficient in this configuration is 33% of that of configuration 1. The numerical results for configurations 4, 6, 8,9, 11, 13 are shown in Fig. 9.

In both of these configurations the heat transferred from channel zone is greater than the one from the diffuser as similar to configuration 1 and 2. For outer (dome and base) zones, however, the heat transferred from dome zone is greater than the one of base zone.

It is seen from Fig. 9b and c that for configurations 4–8 the airtemperature gets higher values near the channel and diffuser zones due to the heat transfer from indoor environment. This causes the increase in buoyancy force and results in the air flow in the upward direction near the TDD wall. When the flow reaches separator plate it losses heat and its temperature drops. The cold air moves downward at the center of the TDD. The heat conducted through the separator plate is transferred to the outdoor environment by the flow in the upper volume. In configuration 6 the

velocities are lower than those of other cases. Due to the low conductivity of acrylic separator plate and the flow patterns, the bottom volume average temperature in this configuration is greater than the average temperature in other configurations. Thus this configuration has the lowest total heat transfer rate among the other configurations. For configurations 9–13 however, the upper volume flow pattern changes. In these configurations the air temperature gets higher values near the channel zone due to the heat transfer from indoor environment and near the separator plate due to the conducted heat. This causes the increase in buoyancy force and results the air flow in the upward direction near the channel zone wall. Simultaneously air cools near the dome and base zone walls due to the heat transfer to the outdoor environment. This causes decrease in buoyancy force and results in air flow in the downward direction near the wall. These two streams mix thereabout insulated zone and moves towards the center of the TDD. Due to these flow patterns total heat transfer rate increases significantly in these configurations.

Q/Q

0


The effect of heat transfer rates on the building thermal load was determined by using simulation method given in Turkish standard TS 825 [23]. In this method, conduction, convection and ventilation heat losses, internal and solar heat gains were taken into consideration. By using this method, monthly averaged heat losses and gains were calculated and the annual heating energy requirement of a building was determined. In this study a building having 3025 m2 floor space was simulated as an example. Thewall, floor and ceiling properties of a building were taken from theTS 825 ( Ufloor = 0.7 W/m2 K, Uceiling = 0.45 W/m2

Kand

Uwall = 0.7 W/m2 K). Simulations were performed for the following cases:

Fig. 8. Overall heat transfer coefficient (UTDD) values for different configurations.

Case-1: Building having no TDD. Case-2: Building having 100 TDDs specified in configuration 1. Case-3: Building having 100 TDDs specified in configuration 6.

a70

DomeZone 60

Base 50Zone

40

30

ChannelZone 20

Base ZoneDome Zone

Diffuser ZoneChannel Zone

z rDiffuserZone

b

10

0Conf.-4 Conf.-6 Conf.-8 Conf.-9 Conf.-11 Conf.-13

Temp.(K)

cVel.(m/s)

Conf.-4 Conf.-6 Conf.-8 Conf.-9 Conf.-11 Conf.-13

Conf.-4 Conf.-6 Conf.-8 Conf.-9 Conf.-11 Conf.-13

Fig. 9. Numerical results for configurations 4, 6, 8, 9, 11, 13. (a) Q/Q0 (b)Temperature (c) Velocity.


Table 2Comparison of simulation results.

Annual heating energy requirement(kWh)

The deviation from case-1 (%)

overall heat transfer coefficients obtained for TDDs are higher than that for ceilings. Although the amount of this increase is too small it can be minimized by appropriate modifications made on the TDDs. These modifications include placing a separator plate into TDD, insulating the outer surfaces of the tubes, and modifying the

Case-1 110,534.97 –Case-2 114,540.74 3.62Case-3 111,450.29 0.83

The comparison of simulation results for the cases is listed inTable 2.

It is seen from Table 2 that the annual heating energy requirements for case-2 and case-3 are 3.62% and 0.83% greater than that of case-1, respectively.

4.3. Discussion

The aim of this study is to determine the effect of the usage of TDDs (one of the structural members) on the building heat loads. The main parameter for the determination of the heat load of the structural member is the overall heat transfer coefficient. This coefficient varies according to the structural member, the heat transfer coefficient value of the various structural members is given in TS 825. However, there is no heat transfer coefficient value for TDDs given in TS 825. This shortcoming has been the starting point of the study.

In this paper a part of the work performed to determine the overall heat transfer values of TDDs manufactured by “Form En- dustri Tesisleri A.S.” is presented. At the beginning of the study the natural convection heat transfer occurring into the TDD having55 cm diameter is examined experimentally. For this purpose, a test system was established and the overall heat transfer coefficient of TDD was determined by using this test system. The experimental results show that the overall heat transfer coefficient for a standard TDD (shown in configuration 1) has been obtained as 3.78 W/m2 K. Then, numerical studies were performed and thermal transmittance coefficients of various TDD configurations were calculated. It is seen from numerical results for a standard TDD (shown in configuration 1) that the numerical results agree very well with the experimental results with an error of 2%. Furthermore the data obtained using this numerical model shows that the heat transfer in TDDs is dependent on the flow pattern. In the next phase of the study, modifications were as investigated in order to reduce heat transfer among them, the applicability of the easiest, and the lowest cost is the insertion of a separator plate into TDD. At this stage of the work, experimental and numerical studies were performed for the TDD shown in configuration 2 in which the separator plate is placed at the top of the channel. The results show that the overall heat transfer coefficient in configuration 2 is 62% of that in configuration 1. The effect of the placement of a separator plate on heat transfer is another issue discussed in this paper. It is seen from the numerical results that the minimum overall heat transfer coeffi-cients (UTDD = 1.22 W/m2 K) obtained for the configuration in whichthe separator plate is placed at the top of the insulated zone (shown in configuration 6). In this case the overall heat transfer coefficient is 33% to that of a standard TDD. The effect of these overall heat transfer coefficients on the building thermal load was analyzed by using simulation method given in Turkish standard TS 825. The simulation results shows that the annual heating energy requirements for case-2 and case-3 are 3.62% and 0.83% greater than that in case-1, respectively.

5. Conclusion

This study indicates that the amount of energy required for heating indoors increases by using TDDs due to the fact that

T. Pirasaci / Journal of Building Engineering 4 (2015) 66 T. Pirasaci / Journal of Building Engineering 4 (2015) dome structure. In this study, only one of them (placing separator plate into TDDs) is discussed for winter conditions. Results shows that this modification significantly decreases heat transfer from TDDs. On the other hand; this study does not include summer conditions, it can be said that the heat transfer also decreases in summer by using a separator plate due to the same convection mechanism. But in summer conditions solar radiation is very im- portant and it must be considered in the analysis. More work is needed to investigate the effect of other modifications on the thermal performance of TDDs and future works should include the radiation effects and the summer conditions.

Acknowledgements

The author wishes to thank the “Form Endustri Tesisleri A.S.” for their contribution to the foundation of the experimental setup and providing the Tubular Daylighting Device product information and specifications.

References

[1] R. Leslie, Capturing the daylight dividend in buildings: why and how?, Build.

Environ. 38 (2003) 381–385.[2] M. Todorovic, J. Kimb, Buildings energy sustainability and health research via

interdisciplinarity and harmony, Energy Build. 47 (2012) 12–18.[3] S. Chandratilake, W. Dias, Sustainability rating systems for buildings: com-

parisons and correlations, Energy 59 (2013) 22–28.

[4] M. Mayhoub, D. Carter, Towards hybrid lighting systems: a review, Light. Res.

Technol. 42 (2010) 51–71.

[5] J. Kim, G. Kim, Overview and new developments in optical daylighting systems for building a healthy indoor environment, Build. Environ. 45 (2010) 256–269.

[6] D. Jenkins, T. Muneer, Modelling light-pipe performances – a natural daylighting solution, Build. Environ. 38 (2003) 965–972.

[7] D. Li, L. Yang, J. Lam, Zero energy buildings and sustainable development implications: a review, Energy 54 (2013) 1–10.

[8] L. Mavromatidis, X. Marsault, H. Lequay, Daylight factor estimation at an earlydesign stage to reduce buildings energy consumption due to artificial lighting: a numerical approach based on Doehlert and Boxebehnken designs, Energy 65 (2014) 488–502.

[9] D. Carter, Developments in tubular daylight guidance systems, Build. Res. Inf.32 (2004) 220–234.

[10] J. Mohelnikova, Tubular light guide evaluation, Build. Environ. 44 (2009)2193–2200.

[11] Y. Su, N. Khana, S. Riffat, O. Garetha, Comparative monitoring and data re- gression of various sized commercial lightpipes, Energy Build. 50 (2012)308–314.

[12] A. McGowan, A. Desjarlais, J. Wright, Simulation and testing of pyramid and barrel vault skylights, ASHRAE Trans. 104 (1998) 832–844.

[13] J. Klems, U-values of flat and domed skylights, ASHRAE Trans. 2000, 106.[14] A. Laouadi, M. Atif, Natural convection heat transfer within multi-layer

domes, Int. J. Heat Mass Transf. 44 (2001) 1973–1981.[15] A. Sartipi, A. Laouadi, D. Naylor, R. Dhib, Convective heat transfer in domed

skylight cavities, J. Build. Perform. Simul. 3 (2010) 269–287.[16] H. Saber, A. Laouadi, Convective heat transfer in hemispherical cavities

with planar inner surfaces, J. ASHRAE Trans. 2011, 117 (Part 2).[17] H. Saber, A. Laouadi, A. Galasiu, C. Arsenault, Convective heat transfer corre-

lations for low-profile spherical cavities with planar bottom surfaces, HVAC&R Res. 19 (2013) 10–23.

[18] R. Moffat, Contributions to the theory of single-sample uncertainty analysis, J.

Fluids Eng. 104 (1982) 250–260.[19] R. Moffat, Using uncertainty analysis in the planning of an experiment, J. Fluids

Eng. 107 (1985) 173–178.

[20] R. Abernethy, R. Benedict, R. Dowdell, Asme measurement uncertainty, J.Fluids Eng. 107 (1985) 161–164.

[21] S. Kline, The purposes of uncertainty analysis, J. Fluids Eng. 107 (1985) 153–160. [22] R. Smith, S. Wehofer, From measurement uncertainty to measurement com-

munications, credibility, and cost control in propulsion ground test facilities, J. Fluids Eng. 107 (1985) 165–172.

[23] TS 825 – Thermal Insulation Requirements for Buildings, Turkish Standart, May 2008.

science direct traducido al spanish

Documents