PhD Oral Exam - Zeinab Khorasani Zadeh, Building Engineering

When studying for a doctoral degree (PhD), candidates submit a thesis that provides a critical review of the current state of knowledge of the thesis subject as well as the student’s own contributions to the subject. The distinguishing criterion of doctoral graduate research is a significant and original contribution to knowledge.

Once accepted, the candidate presents the thesis orally. This oral exam is open to the public.

Abstract

Residential electric heating is a major contributor to winter peak electricity demand in cold climates. Thermostat-based direct load control (DLC) can reduce these peaks through preheating and setbacks. However, fixed control parameters overlook differences in building thermal performance and occupant comfort preferences. Consequently, the same strategy may shift heating demand in one home but cause uncomfortable temperatures in another, prompting occupants to override the imposed settings and reducing intended peak demand reduction. Adapting control parameters to these differences is challenging when building characteristics and occupant preferences are unknown. This thesis develops an adaptive occupant-centric thermostat control algorithm that personalizes control parameters using data available to utilities through smart thermostats, without requiring prior knowledge of building thermal characteristics or occupant preferences. It first analyzes real-world thermostat data to examine household differences in DLC overrides and develop models that predict occupant overrides during DLC. Thermostat adjustment models are then developed using data from multi-zone Quebec homes, complemented by a comparison of centralized and decentralized heating systems. These investigations inform occupant thermostat behaviour representations within an EnergyPlus-Python simulation framework. The framework supports algorithm development and comparison with static DLC across occupant types and building envelopes. Evaluation extends to controlled cooling experiments at the National Research Council Canada’s twin-house facility and heating-season deployment in occupied homes participating in Hydro-Québec’s demand response. Field observations inform refinements, which are subsequently evaluated in simulation. Ultimately, this thesis demonstrates a practical approach to personalized residential DLC using existing thermostat infrastructure, without detailed building models or extensive prior training.