Net ZEB Design Processes and Tools - Modeling, Design, and Optimization of Net-Zero Energy Buildings - page 219

Fig. 4.21 National Renewable Energy Laboratory (NREL) Research
Support Facility (RSF): window with fixed louvers on top half (left) and the
resulting upward-reflected daylight on the ceiling of the office space (right)
In an example by O'Brien (2013), a small 9 m 2 single-occupancy
south-facing office was modeled with a stochastic occupant behavior model.
The occupant was assumed to have randomly varying solar thresholds that
triggered them to close the shade and then reopen it after a varying number
of days (a normal distribution with a mean of 3 days and standard deviation
of 3 days). After 100 simulation runs in EnergyPlus (i.e., Monte Carlo
analysis), a normal probability distribution was fit to the simulated lighting
energy results (right side of Figure 4.22 ) . The simulations were repeated for
a case with a 2 m deep overhang above the window. The results indicate that
thepresenceoftheoverhanggreatlyreducesmeansimulatedenergyuseand
uncertainty (standard deviation) because the direct solar radiation on the
workplane is reduced. Thus, the glare conditions that trigger the occupant
to close their shade are much rarer and as a result, the occupant leaves the
shade open for longer and admits more diffuse daylight.