Master’s Thesis Presentation • ISS4E — Power-Optimal Smart Lighting Control System: Modeling, Implementation and EvaluationExport this event to calendar

Friday, August 16, 2019 1:30 PM EDT

Yerbol Aussat, Master’s candidate
David R. Cheriton School of Computer Science

Lighting load accounts for approximately one third of overall energy consumption in modern office buildings. To reduce this load, we have designed a smart lighting control system that minimizes power consumption, while simultaneously increasing occupant comfort, by dynamically accommodating heterogeneous illuminance requirements as well as changes in occupancy. 

Most current daylight-harvesting lighting systems measure lighting levels at the luminaires or at the walls to deduce illuminance on work surfaces. However, this computation is prone to error, which can potentially result in compromised user comfort. Instead, our system measures illuminance and occupancy directly from sensors located at each work station. It uses sensor readings to dynamically estimate the relationship between the dimming level of each luminaire and the illuminance at each work surface using an unobtrusive calibration process. Subsequently, a linear-programming-based adaptive control algorithm determines power-efficient and comfort-preserving dimming levels for each luminaire. A plug-and-play design lets us seamlessly connect and disconnect system components, such as additional luminaires and sensing modules, even while the system is in use. 

Based on a deployment of our system in the real office environment, we demonstrate that it maintains the desired illuminance at work surface despite environmental fluctuations. We also show, through extensive simulations using 7 months of collected daylight and occupancy data, that our system substantially reduces energy consumption compared even to an occupancy-aware LED lighting system.

Location 
DC - William G. Davis Computer Research Centre
1304
200 University Avenue West

Waterloo, ON N2L 3G1
Canada

S M T W T F S
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
1
2
3
4
5
  1. 2024 (96)
    1. April (19)
    2. March (27)
    3. February (25)
    4. January (25)
  2. 2023 (296)
    1. December (20)
    2. November (28)
    3. October (15)
    4. September (25)
    5. August (30)
    6. July (30)
    7. June (22)
    8. May (23)
    9. April (32)
    10. March (31)
    11. February (18)
    12. January (22)
  3. 2022 (245)
  4. 2021 (210)
  5. 2020 (217)
  6. 2019 (255)
  7. 2018 (217)
  8. 2017 (36)
  9. 2016 (21)
  10. 2015 (36)
  11. 2014 (33)
  12. 2013 (23)
  13. 2012 (4)
  14. 2011 (1)
  15. 2010 (1)
  16. 2009 (1)
  17. 2008 (1)