Artificial Intelligence in Performance-Driven Design

3A Hybrid Physics‐Based Machine Learning Approach for Integrated Energy and Exposure Modeling

Mehdi Ashayeri¹ and Narjes Abbasabadi²

¹ School of Architecture, College of Arts and Media, Southern Illinois University Carbondale, Carbondale, IL, USA

² Department of Architecture, School of Architecture, College of Built Environments, University of Washington, Seattle, WA, USA

Introduction

According to the National Human Activity Pattern Survey (NHAPS), in the United States, people spend about 87% of their time indoors (Klepeis et al. 2001). Indoor spaces are required to ensure health and comfort for the occupants, ensuring sustainable built‐environment goals. Heating, ventilating, and air conditioning (HVAC) systems are used to provide thermally comfortable conditions and maintain “better” indoor air quality (IAQ) for occupant health (Dutton et al. 2013). According to the U.S. Environmental Protection Agency (EPA), IAQ “refers to the air quality within and around buildings and structures, especially as it relates to the health and comfort of building occupants” (US EPA 2014). The HVAC systems, however, contribute to 39% of the overall building energy consumption (Annual Energy Review 2011 2012); and cooling and ventilation systems account for 31% of the end‐use energy in the U.S. commercial building stock (Use of Energy Explained‐Energy Use in Commercial Buildings 2018). Natural ventilation (NV), as a passive strategy, which often is used to reduce cooling and ventilation energy ...