CFD and Deep Learning Based Natural Ventilation Analysis in Buildings
Bibliographic record
Abstract
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.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".