Modeling and Design of Net ZEBs as Integrated Energy Systems - Modeling, Design, and Optimization of Net-Zero Energy Buildings - page 58

unknown parameters ( I ph , I o , R s , R sh , and a ) can be found in topics such as
“Solar Engineering of Thermal Processes” (Duffie and Beckman, 2006).
Modeling the Thermal Performance
Predicting the PV cell temperature is essential as it affects the module
electrical efficiency. The European Union project PVSAT-2 (2006), aiming
at establishing the performance of PV system operation based on satellite
data, used the following relation to estimate the PV cell operating
temperature:
(2.29)
where C isanempiricallydeterminedcoefficient(0.058foraroof-integrated
installation) (m 2 KW −1 ) (Drews et al. , 2007), G is total incident solar
radiation (Wm −2 ), and T a is ambient temperature (K).
King, Boyson, and Kratochvil (2004) developed a more accurate (less than
3% effect on module power output) empirically based thermal model to
predict PV cell temperature that takes into account the effect of wind speed
and the temperature gradient between the backsheet of the module and the
actual PV cells:
(2.30)
where is the reference total incident solar radiation (1000 Wm −2 ), V
is the wind speed measured at standard height of 10 m (ms −1 ), C 1 is the
empirically determined coefficient expressing the upper limit for module
temperatureatlowwindspeedsandhighsolarirradiance(−2.98foraglass/
cell/glass, close roof mount module and −2.81 for a glass/cell/polymer,
insulated-back module), C 2 is the empirically determined coefficient
expressing the rate at which module temperature drops as wind speed
increases (−0.0471 for a glass/cell/glass, close roof mount module and
−0.0455 for a glass/cell/polymer, insulated-back module), and Δ T is the
temperature difference between the cells and the module backsheet, under
1000 Wm −2 (1 K for a glass/cell/glass, close roof mount module and 0 K for
a glass/cell/polymer, insulated-back module).
The temperature of PV cells can be estimated using finite difference models.
This approach is useful when empirical coefficients for the specific module