Evaluation of blinds control techniques for daylight and visual comfort in complex real-world conditions

The presence of sensor networks to monitor environmental conditions and the automation of blinds and lighting systems controls is now commonplace in buildings, especially public ones with a high number of occupants. However, implementing control algorithms that are sufficiently reactive to variable sky conditions and that actually meet occupants’ needs is still a challenge. In the present study, we investigate and compare advanced and simple control algorithms developed for a variable occupancy, open space, small sized conference venue. Operation and performance resulting from an optimized approach are assumed to be the benchmark strategy, and two other control algorithms of varying complexity are compared with it. Results show that the optimized control strategy performs best overall, but only marginally compared to the other two strategies. It performed especially well in meeting glare protection requirements, as a glare-related parameter was embedded into its objective function, but it also led to erratic movements of the blind slats’ tilt and it required significantly higher computation times than rule-based control strategies. These two factors make it impossible to implement such strategy as it is in the real building, and indicate that a practical control implementation can be more effective than an optimal one.