Given people's significant time spent indoors, ensuring good indoor air quality (IAQ) is essential because it significantly influences occupants' health and productivity. Office buildings consume about 50% of commercial building energy and 18% of total building stock, with HVAC systems contributing around 40% of the energy consumption. Despite the advent of low-cost smart building sensors, there is a lack of guidelines for optimal IAQ monitoring strategies particularly in offices with dynamically changing occupancy. This thesis investigates three research topics: 1) proxy methods for inhalation exposure assessment, 2) optimal air pollution sensor placement which captures inhalation exposures, and 3) sets of indicators for air inhalation exposure and occupancy detection in office environments.
Chapter 3 proposes proxy methods for detecting personal inhalation exposures to carbon dioxide (CO2) and particulate matter (PM) in simulated office settings with dynamically changing occupancy. Three proxy sensing techniques were compared with the concurrent breathing zone measurements: stationary IAQ monitoring, wearable wristbands for physiological monitoring, and passive infrared (PIR) sensors for human presence detection. Combining three proxy techniques improved the CO2 exposure detection by twofold compared to solely using a stationary IAQ monitor. Stationary PM monitors near the ventilation exhaust accurately estimated PM exposure, while CO2 measurements at the front edge of the desk showed moderate accuracy for CO2 exposure detection.
Chapter 4 investigates optimal sensor placement for detecting inhalation exposure in static and dynamic simulated office environments. It identifies suitable locations for accurate exposure estimation, considering occupancy dynamics. The findings show that differentiating between static/dynamic occupancy and sitting/standing activities enhanced the accuracy of exposure detection. Variables such as proximity of sensors to occupants and ventilation rate/strategy played significant roles in improving personal exposure detection. Desk- and wall-mounted CO2 sensors, along with a ceiling-mounted PM sensor, provided the most accurate exposure detection.
Chapter 5 explores indicators for exposure and occupancy detection in two real office buildings in the western part of Switzerland. The method used a combination of stationary and wearable sensors, along with Decision Tree and correlation analyses. Occupancy strongly influenced air pollution gradients in different office spaces, with higher PM10 levels during lunch/coffee activities. Desk-mounted CO2 sensors effectively detected CO2, PM2.5, and PM10 exposures in open-plan offices. CO2 levels at the sidewall represented prolonged occupancy, while desk-mounted PM10 sensors captured transient occupancy. A single CO2 sensor proved to be a cost-effective solution for both CO2, PM2.5 and PM10 exposure and occupancy detection. Air pollution data demonstrated up to 4× higher predictive power in detecting exposures and occupancy compared to indoor climate data.
The thesis proposes optimizing solutions for exposure and occupancy detection with smart building sensors under various office setups and occupancy scenarios. The findings could find application in enhancing IAQ management and occupant-centric HVAC control through integrated smart monitoring techniques that can be used in real-life occupancy conditions.