Temperature profile at top of panel

Bottom temperature profile Figure 4. Temperature profiles (horizontal) at top and bottom of cavity for the case of Fig.2.

One-dimensional models for the two double fagade concepts described above have been developed by Charron and Athienitis [4]. Here, a simplified version of these models is employed for the case of Figure 2 in order to study the impact of major parameters such as convective heat transfer coefficient. Considering the fagade with PV as exterior layer we may represent it with the thermal network model shown in Figure 5.

Figure 5. Thermal network model of fagade with exterior PV (assuming isothermal surfaces); node b indicates the back panel interior surface.

PV

The mean air temperature Tma is determined from a differential analysis which finds the air temperature as a function of vertical distance x. It is assumed that the air speed is constant, that is, air it is drawn into the window by a fan in the HVAC system fresh air intake. The actual air temperature T(x) is then used to determine the Tma. This is then employed to find the correct values of Tpv and Tb which are utilized to fine tune the calculations. Considering an element dx in the vertical direction, we have:

M-c-p-dT= W•dx-h-(Tpv — T) + W-dx-h-(Tb — t) (!)

where M = flow rate = V * A (V is average velocity and A is cross-sectional area) and W is width of fagade. Note that the convective coefficient h is an average for both cavity surfaces (in reality it will generally be higher on the hotter surface).

Note that in this simple model we assume equal convective heat transfer coefficient h for both cavity surfaces. The following ordinary differential equation is obtained:

(2)

a-—T + 2T= Tb + Tpv dx P

M-c-p

a :=

with Wh

T(x) ■

To —

(T b + T pv)

2

(3)

e

— X-2

T pv + T b 2

An exponential variation is obtained for the air temperature as follows:

Tb :

T pv

(4)

(5)

TmaU b + T rU 3 + T pvU r U3+Ub+Ur

U o’T o + U a — Tma + Ur-T b + Spv Uo+Ua+Ur

The PV and back panel temperatures are obtained as:

where U represents conductance between the various nodes (Uo= A ho, Ur= A hr,

Ua= Ub= A h, and U3 is negligible).

The average surface temperatures predicted with this model for various values of the convective heat transfer coefficient are compared in Table 1. Note that there is significant uncertainty about the value of this coefficient and this team is currently performing CFD studies on this topic. Nevertheless, simple models such as the one presented above are instrumental in studying the fundamentals of the problem. A value of 14 W/m2K was employed for the exterior film coefficient ho (other measurements also confirm this value) and 4.0 for the radiation heat transfer coefficient hr.

As can be seen from Table 1, a value of about 8 W/m2K for h gives reasonably accurate temperature predictions for the air and for the two surfaces. Note that the accuracy of the average velocity measurement is about 5%. The value of the convective heat transfer

coefficient is relatively high because of the short height of the PV panel (about 1m) and as expected h is high near the leading edge of the boundary layer forming on the PV panel.

Table 1. Predictions of 1-D model as a function of convective heat transfer coefficient Incident solar radiation is 989 W/m2, Jan. 26, 2004 Convective heat transfer coefficient h, W/(m2 0C) Average temperature of PV panel, 0C Average air temperature rise between bottom and top of PV, 0C Thermal efficiency 5 23.3 3.8 28% 6 21.4 4.2 32% 7 19.8 4.6 34% 8 18.2 4.9 36% 9 17 5.1 38% 10 15.7 5.4 40% Experimental results 15.5 4.9

Conclusion

The results of an experimental study and a simple analytical model for a double skin fagade with integrated photovoltaic panels are presented and analyzed. Air enters from the bottom part of the fagade through an intake, gets heated as it flows upwards driven by buoyancy and a fan, and finally enters the HVAC system. During the winter, fresh air increases the efficiency of the photovoltaic panels as it flows around them while at the same time it is preheated. Experimental results show that combined thermal-electric efficiency of the system could easily exceed 70% with airflow on both sides of the PV panel.

A major result of the study is estimation of the impact of the convective heat transfer coefficient. For velocities of about 0.6 m/s, an average coefficient of 8 W/m2K (over a height of 1 m) was found to give good agreement with experimental measurements for air temperature rise and for the average temperature of the panels.

Acknowledgements

The financial and in-kind support of this project from NSERC, ATS, Dept of Natural Resources of Quebec, and CETC-Varennes is gratefully acknowledged.

References

1. IEA 1999, Workshop on PV/Thermal Solar Systems, Amersfoot, Netherlands, 17-18 Sep.

2. Lloret et al. (1995) The Mataro public library: a 53 KWp grid connected building with integrated PV — thermal multifunctional modules. 13th European PV Solar Energy Conference, Nice, France, pp. 490 — 493.

3. D. Infield, L. Mei, U. Eicker, "Thermal performance estimation for ventilated PV fagades”, Solar Energy, Vol. 76, pp. 93-98

4. Charron R. and Athienitis A. K., 2003, Optimization of the Performance of PV-Integrated Double-Fagades, Proc. of International Solar Energy Society World Congress, Goteborg, June.