Abstract
Due to computational barriers of computational fluid dynamics (CFD) models, they cannot be used for tasks such as (near) real-time simulations. Reduced-order model (ROM) can be used as an alternative to CFD since it can approximate the results in a fraction of the CFD simulation time. The present article generates a data-driven ROM, using convolutional autoencoders (CAEs) and long short-term memory (LSTM) networks, to reconstruct the turbulent flow field within a simplified urban area. Furthermore, the effect of the kernel size on capturing spatial information is investigated. The results indicate that, although the model has some deficiencies in the flow field reconstruction in high-gradients regions, the model's overall performance is acceptable. Moreover, it is shown that the kernel size has a negligible impact on the model performance for the present model and dataset.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Blocken B (2018) LES over RANS in building simulation for outdoor and indoor applications: a foregone conclusion? 11(5)
Ding S, Yang R (2021) Reduced-order modelling of urban wind environment and gaseous pollutants dispersion in an urban-scale street canyon. J Saf Sci Resil 2(4):238–245. https://doi.org/10.1016/j.jnlssr.2021.09.001
Germano M, Piomelli U, Moin P, Cabot WH (1991) A dynamic subgrid-scale eddy viscosity model. Phys Fluids A 3:1760–1765. https://doi.org/10.1063/1.857955
Gonzalez FJ, Balajewicz M (2018) Deep convolutional recurrent autoencoders for learning low-dimensional feature dynamics of fluid systems. arXiv:1808.01346 [math.DS]
Kingma DP, Ba JL (2015) Adam: a method for stochastic optimization. In: 3rd international conference on learning representation, ICLR 2015—conference track proceedings, pp 1–15
Leung DYC (2015) Outdoor-indoor air pollution in urban environment: challenges and opportunity. Front Environ Sci 2:1–7. https://doi.org/10.3389/fenvs.2014.00069
Li Z et al (2021) Review on pollutant dispersion in urban areas-part A: Effects of mechanical factors and urban morphology. Build Environ 190:107534. https://doi.org/10.1016/j.buildenv.2020.107534
Lilly DK (1992) A proposed modification of the Germano subgrid-scale closure method. Phys Fluids A 4:633–635. https://doi.org/10.1063/1.858280
Masoumi-Verki S, Gholamalipour P, Haghighat F, Eicker U (2021) Embedded LES of thermal stratification effects on the airflow and concentration fields around an isolated high-rise building: spectral and POD analyses. Build Environ 206:108388. https://doi.org/10.1016/j.buildenv.2021.108388
Masoumi-Verki S, Haghighat F, Eicker U (2022) A review of advances towards efficient reduced-order models (ROM) for predicting urban airflow and pollutant dispersion. Build Environ 216:108966. https://doi.org/10.1016/j.buildenv.2022.108966
Mücke NT, Boht SM, Oosterlee CW (2021) Reduced order modeling for parameterized time-dependent PDEs using spatially and memory aware deep learning. J Comput Sci 53:101408
Nakamura T, Fukami K, Hasegawa K, Nabae Y, Fukagata K (2021) Convolutional neural network and long short-term memory based reduced order surrogate for minimal turbulent channel flow. Phys Fluids 33:025116. https://doi.org/10.1063/5.0039845
Shih T-H, Liou WW, Shabbir A, Yang Z, Zhu J (1995) A new k-ϵ eddy viscosity model for high reynolds number turbulent flows. Comput Fluids 24(3):227–238. https://doi.org/10.1016/0045-7930(94)00032-T
W. H. O. (WHO) (2021) www.who.int/news-room/fact-sheets/detail/ambient-(outdoor)-air-quality-and-health
Wang Z, Xiao D, Fang F, Govindan R, Pain CC, Guo Y (2018) Model identification of reduced order fluid dynamics systems using deep learning. Int J Numer Methods Fluids 86(4):255–268. https://doi.org/10.1002/fld.4416
Wu P, Gong S, Pan K, Qiu F, Feng W, Pain C (2021) Reduced order model using convolutional auto-encoder with self-attention. Phys Fluids 33(7). https://doi.org/10.1063/5.0051155
Xiang S et al (2021a) Fast simulation of high resolution urban wind fields at city scale. Urban Clim 39:100941. https://doi.org/10.1016/j.uclim.2021.100941
Xiang S et al (2021b) Non-intrusive reduced order model of urban airflow with dynamic boundary conditions. Build Environ 187:107397. https://doi.org/10.1016/j.buildenv.2020.107397
Yoshie R (2016) Wind tunnel experiment and large eddy simulation of pollutant/thermal dispersion in non-isothermal turbulent boundary layer. In: Advanced environmental wind engineering, pp 167–196
Zhang Y, Gu Z, Yu CW (2020) Impact factors on airflow and pollutant dispersion in urban street canyons and comprehensive simulations: a review. Curr Pollut Reports 6(4):425–439. https://doi.org/10.1007/s40726-020-00166-0
Acknowledgements
The authors would like to express their gratitude to Concordia University—Canada for the support through the Concordia Research Chair—Energy and Environment, and Canada Excellence Research Chairs Program.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Masoumi-Verki, S., Haghighat, F., Eicker, U. (2023). Data-Driven Reduced-Order Model for Urban Airflow Prediction. In: Wang, L.L., et al. Proceedings of the 5th International Conference on Building Energy and Environment. COBEE 2022. Environmental Science and Engineering. Springer, Singapore. https://doi.org/10.1007/978-981-19-9822-5_324
Download citation
DOI: https://doi.org/10.1007/978-981-19-9822-5_324
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-19-9821-8
Online ISBN: 978-981-19-9822-5
eBook Packages: EngineeringEngineering (R0)