Understanding occupant thermostat preferences and thermostat setpoint override behavior is critical for exploiting thermostat setpoints or HVAC systems for energy flexibility and demand response potential. This study models occupant thermostat overrides in residential buildings during the heating season in Quebec, Canada. Two distinct occupant types, 'Average' and 'Tolerant', are identified based on their thermostat setpoint preferences. Discrete-time Markov logistic regression models are developed to predict the probability of thermostat setpoint overrides during 'Home', 'Sleep', and 'Work' hours for both occupant types. Random forest models are employed to classify the magnitude of the setpoint changes as either small (less than 0.5 °C) or large (greater than 0.5 °C). Using these models, we are thus able to estimate the probability of a setpoint override and the magnitude of the setpoint override for two distinct occupant types in Quebec. Results indicate good model performance with balanced accuracy, recall, and precision. However, limitations include data availability, model assumptions, and the need for more comprehensive occupancy data. These models offer potential applications in demand response and HVAC control strategies for residential buildings.

Building-to-grid (B2G) integration research aims to use buildings as flexible power demand resources to provide grid services. Occupant behaviors in residential homes, one of the major building types, significantly influence power demands. This paper develops a novel modeling framework for residential homes’ occupant behaviors supporting B2G integration research. This high-fidelity modeling framework is built based on well-accepted models and controls. In addition, the proposed framework provides an interface to connect standardized power system tools with building models. It has the flexibility of scaling up models and controls from a single building to communities and urban districts.

Smart thermostats are a low-complexity and widely available technology for reducing HVAC energy use in homes. But how well do they actually work?

Most existing methods of evaluating smart thermostats rely on data collected from installations. This requirement makes it difficult to evaluate thermostats that have limited installations or even no installations. For instance, the EPA requires 12 months of data from 1,250 qualifying installations before granting the ENERGY STAR label. Data-driven methods are also unable to directly compare multiple thermostats in controlled experiments because setpoint time series cannot be collected for multiple thermostats operating in the same home during the same period.

Recent advances in energy simulation can be used to overcome these shortcomings. In this work, we leverage these advances to create a simulation-driven smart-thermostat evaluation framework and use it to evaluate simple thermostats as well as a generic smart thermostat algorithm for 52 representative homes with various types of HVAC equipment, extending previous results in this area.

Accurately representing buildings’ occupancy schedules has been crucial in understanding the energy flexibility of buildings as well as improving the allocation of building services and resources. However, the limited availability of data to model occupancy schedules for specific buildings has been challenging the development of an accurate prediction model. This study evaluates a parameter-based transfer learning scheme developed for time series prediction of occupancy schedules with limited data. The study focuses on an institutional building in Quebec, Canada as a case study. The results showed that transferring the knowledge of the pre-trained time series model has not only reduced the fitting computational power but also reduced the model’s training requirements while maintaining a mean absolute error of 3 occupants.

Investigating socio-cultural induced occupant behavior and its impact on indoor thermal comfort is essential in dense low-income tenements in India, owing to financial and space restraints. This study utilized household surveys followed by sensor-based field measurements to explore the occupant behavior-driven indoor temperature trends within slum rehabilitation housing in Mumbai.

The findings underscore the use of ceiling fans and windows as primary strategies for adaptive comfort. Regression analysis validated an association between air temperature, fan, and cooking. However, fans dominate natural ventilation, as privacy concerns and architectural design limit window utilization. The study offers valuable insights for enhancing thermal comfort, in the context of energy-efficient design.