Master of Engineering Science
Civil and Environmental Engineering
Bitsuamlak, Girma T.
Capretz, Miriam A.M.
Natural ventilation is crucial for sustainable buildings and is also a promising solution for addressing indoor air quality (IAQ) issues such as those related to COVID-19. This thesis examines the efficacy of wind-driven cross-ventilation for a low-rise residential building with complex geometry and internal partitions typical of common constructions using computational fluid dynamics (CFD) simulations. Different wind speeds and directions with varying partition and window configurations are analyzed, as well as surrounding buildings consistent with Canadian suburban neighbourhoods. While CFD solvers are effective in predicting natural ventilation, they are limited by processing time, hardware and storage requirements, and specialized knowledge, which limits the number of designs tested and leads to suboptimal solutions. Therefore, user-friendly deep learning models are also developed to efficiently predict the velocity field within a cross-ventilated building, using both a Vanilla U‑Net and a U-Net with an attention mechanism for the neural network architectures. The models obtain training data from CFD simulations performed on a generic building with multiple opening sizes and impacts from different wind directions. The results show that partition walls block airflow and create dead zones, but when openings are introduced on partition walls to form a network of openings, IAQ is significantly improved, especially in rooms that previously only had a doorway opening. Additionally, surrounding buildings should not be neglected when accounting for IAQ, as the air changes per hour (ACH) can be reduced by more than half, leading to a significant increase in the local mean age of air (MAA) (up to 215%) for the entire building. Furthermore, both deep learning models generate velocity contours much faster than CFD solvers while only sacrificing a small amount of error. However, the Vanilla U-Net model is recommended as it had superior performance in both qualitative and quantitative analyses.
Summary for Lay Audience
The production, transportation, and use of nonrenewable energy have a negative impact on the environment. The building sector is a significant consumer of energy, largely due to the heating, ventilation, and air-conditioning (HVAC) systems. Natural ventilation is a process that can exchange air between the interior and exterior of a building, reducing energy consumption by using exterior temperatures to naturally lower the interior temperature, and bringing in fresh air while dispersing harmful pollutants. This process can greatly improve indoor air quality (IAQ) and is the focus of this thesis.
This thesis tests multiple building designs, wind factors, and surroundings through computer simulations to predict airflow and determine how these parameters affect IAQ. The implementation of a network of small interior openings greatly improved IAQ in rooms that previously only had one air passageway. It is also critical to consider surrounding buildings since they can significantly reduce IAQ compared to an isolated building.
In addition to testing building designs, this thesis develops deep learning models, a type of artificial intelligence, to improve the design process of cross-ventilated buildings and enhance occupant IAQ. Compared to traditional natural ventilation approaches, deep learning is faster, easier to use, and requires minimal equipment. The deep learning models are given the velocity field of many different buildings and learn their patterns, allowing them to approximate airflow. These models accurately predict the velocity field of cross-ventilated buildings within milliseconds, making them orders of magnitude faster than traditional approaches.
Vandewiel, Matthew R., "CFD and Deep Learning Based Natural Ventilation Analysis in Buildings" (2023). Electronic Thesis and Dissertation Repository. 9189.
Available for download on Wednesday, May 01, 2024