Application of Artificial Neural Networks (ANN) for modeling Personalized Ventilation Systems (PVS) using CFD data
DOI:
https://doi.org/10.4025/actascitechnol.v48i1.74146Keywords:
Machine Learning; Computational Fluid Dynamics; Air Conditioning; Predictive model; Indoor airflow.Abstract
This study investigates the application of Artificial Neural Networks (ANNs) for modeling Personalized Ventilation Systems (PVS) using data from Computational Fluid Dynamics (CFD) simulations. In recent years, machine-learning techniques like ANNs have been increasingly used to accelerate CFD analysis and improve the accuracy of temperature and airflow velocity predictions in indoor environments. The methodology involved conducting twelve CFD simulations in a three-dimensional space, followed by data filtering and normalization to train and test the neural network. The room was composed of two individuals, positioned side by side, both seated and receiving air from a ceiling supply system. Both individuals were modeled to maintain a constant surface temperature while also transferring heat to the environment. The quality of the results were analyzed by comparing the neural network outputs with data that had been omitted from the network. The results demonstrated the effectiveness of the model, with average errors ranging from 1% to 3% and maximum errors between 6% and 15%. This approach significantly reduces the computational time required for traditional CFD simulations while maintaining high accuracy, offering promising prospects for optimizing PVS performance in various indoor settings. The use of machine learning makes the analysis and design of personalized ventilation systems faster and more efficient, with practical applications in offices, classrooms, and residential spaces.
Downloads
References
Jing, G., Ning, C., Qin, J., Ding, X., Duan, P., Liu, H., & Sang, H. (2023). Physics-guided framework of neural network for fast full-field temperature prediction of indoor environment. Journal of Building Engineering, 68(1), 106054. https://doi.org/10.1016/j.jobe.2023.106054
Katramiz, E., Ghaddar, N., Ghali, K., Al-Assaad, D., & Ghani, S. (2021). Effect of individually controlled personalized ventilation on cross-contamination due to respiratory activities. Building and Environment, 194, 107719. https://doi.org/10.1016/j.buildenv.2021.107719
Kim, M., & Park, H.-J. (2023). Application of artificial neural networks using sequential prediction approach in indoor airflow prediction. Journal of Building Engineering, 69, 106319. https://doi.org/10.1016/j.jobe.2023.106319
Li, L., Zhang, Y., Fung, J. C. H., Qu, H., & Lau, A. K. H. (2022). A coupled computational fluid dynamics and back-propagation neural network-based particle swarm optimizer algorithm for predicting and optimizing indoor air quality. Building and Environment, 207(B), 108533. https://doi.org/10.1016/j.buildenv.2021.108533
Rissetto, R., Schweiker, M., & Wagner, A. (2021). Personalized ceiling fans: Effects of air motion, air direction and personal control on thermal comfort. Energy and Buildings, 235, 110721. https://doi.org/10.1016/j.enbuild.2021.110721
Saiyad, A., Patel, A., Fulpagare, Y., & Bhargav, A. (2021). Predictive modeling of thermal parameters inside the raised floor plenum data center using artificial neural networks. Journal of Building Engineering, 42, 102397. https://doi.org/10.1016/j.jobe.2021.102397
Tian, X., Cheng, Y., & Lin, Z. (2022). Modelling indoor environment indicators using artificial neural network in the stratified environments. Building and Environment, 208, 108581. https://doi.org/10.1016/j.buildenv.2021.108581
Xu, C., Wei, X., Liu, L., Su, L., Liu, W., Wang, Y., & Nielsen, P. V. (2020). Effects of personalized ventilation interventions on airborne infection risk and transmission between occupants. Building and Environment, 180, 107008. https://doi.org/10.1016/j.buildenv.2020.107008
Xu, J., Fu, S., & Chao, C. Y. H. (2021). Performance of airflow distance from personalized ventilation on personal exposure to airborne droplets from different orientations. Indoor and Built Environment, 30(10), 1643–1653. https://doi.org/10.1177/1420326X20951245
Zhou, Q., & Ooka, R. (2021a). Influence of data preprocessing on neural network performance for reproducing CFD simulations of non-isothermal indoor airflow distribution. Energy and Buildings, 230, 110525. https://doi.org/10.1016/j.enbuild.2020.110525
Zhou, Q., & Ooka, R. (2021b). Neural network for indoor airflow prediction with CFD database. Journal of Physics, 2069, 12154. https://doi.org/10.1088/1742-6596/2069/1/012154
Zhou, Q., & Ooka, R. (2021c). Performance of neural network for indoor airflow prediction: Sensitivity towards weight initialization. Energy and Buildings, 246, 111106. https://doi.org/10.1016/j.enbuild.2021.111106
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Marlon Breno Amora Ribeiro, Álvaro Messias Bigonha Tibiriçá, André Luiz de Freitas Coelho, Júlio César Costa Campos, Henrique Márcio Pereira Rosa (Autor)
This work is licensed under a Creative Commons Attribution 4.0 International License.
DECLARATION OF ORIGINALITY AND COPYRIGHTS
I Declare that current article is original and has not been submitted for publication, in part or in whole, to any other national or international journal.
The copyrights belong exclusively to the authors. Published content is licensed under Creative Commons Attribution 4.0 (CC BY 4.0) guidelines, which allows sharing (copy and distribution of the material in any medium or format) and adaptation (remix, transform, and build upon the material) for any purpose, even commercially, under the terms of attribution.
Read this link for further information on how to use CC BY 4.0 properly.
8.png){kind=logo}
