Integrating Indoor Air Quality and Occupancy Data for Optimized Operation of HVAC Systems

Heating, ventilation, and air conditioning (HVAC) systems account for about 40% of the energy consumed in buildings. However, traditional systems are still unable to dynamically adjust ventilation rates based on real-time indoor air quality (IAQ) metrics and occupancy data. This dissertation discusses the need for HVAC control strategies that dynamically manage ventilation systems to enhance IAQ and energy efficiency. Through a series of studies, this research evaluates the limitations of using carbon dioxide (CO2) as the only indicator of IAQ and explores the inclusion of total volatile organic compounds (TVOC) for a more comprehensive IAQ assessment. Predictive models, including statistical and deep learning approaches, are developed to forecast pollutant levels and occupancy, enabling real-time adjustments to ventilation rates. A four-month field experiment shows that dynamic, demand-driven HVAC operation reduces energy consumption and improves air quality compared to traditional scheduled systems. Additionally, two novel indices were developed to assess the performance loss associated with IAQ and thermal comfort conditions on occupants. The IAQ index, incorporating CO2 and TVOC levels, and the thermal comfort index, using temperature and humidity data, were evaluated during both scheduled and dynamic HVAC operations. Results showed that dynamic operation reduced these indices, indicating enhanced occupant comfort and well-being under dynamic ventilation. This research demonstrates that integrating TVOC, advanced occupancy detection, and predictive models into HVAC management strategies enables a more comprehensive approach to IAQ and energy optimization. Dynamic, demand-driven HVAC systems have the potential to create healthier indoor environments, reduce energy consumption, and support sustainable building practices by adapting ventilation to the specific needs of each environment.