Abstract
Prediction of indoor airflow distribution often relies on high-fidelity, computationally intensive computational fluid dynamics (CFD) simulations. Artificial intelligence (AI) models trained by CFD data can be used for fast and accurate prediction of indoor airflow, but current methods have limitations, such as only predicting limited outputs rather than the entire flow field. Furthermore, conventional AI models are not always designed to predict different outputs based on a continuous input range, and instead make predictions for one or a few discrete inputs. This work addresses these gaps using a conditional generative adversarial network (CGAN) model approach, which is inspired by current state-of-the-art AI for synthetic image generation. We create a new Boundary Condition CGAN (BC-CGAN) model by extending the original CGAN model to generate 2D airflow distribution images based on a continuous input parameter, such as a boundary condition. Additionally, we design a novel feature-driven algorithm to strategically generate training data, with the goal of minimizing the amount of computationally expensive data while ensuring training quality of the AI model. The BC-CGAN model is evaluated for two benchmark airflow cases: an isothermal lid-driven cavity flow and a non-isothermal mixed convection flow with a heated box. We also investigate the performance of the BC-CGAN models when training is stopped based on different levels of validation error criteria. The results show that the trained BC-CGAN model can predict the 2D distribution of velocity and temperature with less than 5% relative error and up to about 75,000 times faster when compared to reference CFD simulations. The proposed feature-driven algorithm shows potential for reducing the amount of data and epochs required to train the AI models while maintaining prediction accuracy, particularly when the flow changes non-linearly with respect to an input.
Article PDF
Similar content being viewed by others
Avoid common mistakes on your manuscript.
References
Albensoeder S, Kuhlmann HC (2005). Accurate three-dimensional lid-driven cavity flow. Journal of Computational Physics, 206: 536–558.
Antipov G, Baccouche M, Dugelay JL (2017). Face aging with conditional generative adversarial networks. In: Proceedings of 2017 IEEE International Conference on Image Processing (ICIP), Beijing, China.
Barth R, Hemming J, Van Henten EJ (2020). Optimising realism of synthetic images using cycle generative adversarial networks for improved part segmentation. Computers and Electronics in Agriculture, 173: 105378.
Blohm C, Kuhlmann HC (2002). The two-sided lid-driven cavity: Experiments on stationary and time-dependent flows. Journal of Fluid Mechanics, 450: 67–95.
Burggraf OR (1966). Analytical and numerical studies of the structure of steady separated flows. Journal of Fluid Mechanics, 24: 113–151.
Cao S (2019). Challenges of using CFD simulation for the design and online control of ventilation systems. Indoor and Built Environment, 28: 3–6.
Castellini JE, Faulkner CA, Zuo W, et al. (2022). Assessing the use of portable air cleaners for reducing exposure to airborne diseases in a conference room with thermal stratification. Building and Environment, 207: 108441.
Chen D, Gao X, Xu C, et al. (2020). FlowGAN: a conditional generative adversarial network for flow prediction in various conditions. In: Proceedings of 2020 IEEE 32nd International Conference on Tools with Artificial Intelligence (ICTAI), Baltimore, MD, USA.
Chiang TP, Sheu WH, Hwang RR (1998). Effect of Reynolds number on the eddy structure in a lid-driven cavity. International Journal for Numerical Methods in Fluids, 26: 557–579.
Cirillo MD, Abramian D, Eklund A (2021). Vox2Vox: 3D-GAN for brain tumour segmentation, In: Proceedings of International MICCAI Brainlesion Workshop.
El-Kaddoury M, Mahmoudi A, Himmi MM (2019). Deep generative models for image generation: A practical comparison between variational autoencoders and generative adversarial networks, In: Proceedings of International Conference on Mobile, Secure, and Programmable Networking.
Gangisetti K, Claridge DE, Srebric J, et al. (2016). Influence of reduced VAV flow settings on indoor thermal comfort in an office space. Building Simulation, 9: 101–111.
Ghia U, Ghia KN, Shin CT (1982). High-Re solutions for incompressible flow using the Navier-Stokes equations and a multigrid method. Journal of Computational Physics, 48: 387–411.
Gonog L, Zhou Y (2019). A review: Generative adversarial networks. In: Proceedings of the 14th IEEE Conference on Industrial Electronics and Applications (ICIEA), Xi’an, China.
Goodfellow IJ, Pouget-Abadie J, Mirza M, et al. (2014). Generative adversarial nets. In: Advances in Neural Information Processing Systems 27 (NIPS 2014).
Han X, Tian W, VanGilder J, et al. (2021). An open source fast fluid dynamics model for data center thermal management. Energy and Buildings, 230: 110599.
Iqbal T, Ali H (2018). Generative adversarial network for medical images (MI-GAN). Journal of Medical Systems, 42: 231.
Jain LC, Seera M, Lim CP, et al. (2014). A review of online learning in supervised neural networks. Neural Computing and Applications, 25: 491–509.
Jayaraman B, Finlayson EU, Sohn MD, et al. (2006). Tracer gas transport under mixed convection conditions in an experimental atrium: comparison between experiments and CFD predictions. Atmospheric Environment, 40: 5236–5250.
Khan MAI, Delbosc N, Noakes CJ, et al. (2015). Real-time flow simulation of indoor environments using lattice Boltzmann method. Building Simulation, 8: 405–414.
Kingma DP, Ba J (2014). Adam: A method for stochastic optimization. arXiv: 1412.6980.
Kong X, Chang Y, Li N, et al. (2022). Comparison study of thermal comfort and energy saving under eight different ventilation modes for space heating. Building Simulation, 15: 1323–1337.
Kuhlmann HC, Wanschura M, Rath HJ (1997). Flow in two-sided lid-driven cavities: non-uniqueness, instabilities, and cellular structures. Journal of Fluid Mechanics, 336: 267–299.
Li Y, Nielsen PV (2011). CFD and ventilation research. Indoor Air, 21: 442–453.
Mirza M, Osindero S (2014). Conditional generative adversarial nets. arXiv: 1411.1784.
Mohamadi F, Fazeli A (2022). A review on applications of CFD modeling in COVID-19 pandemic. Archives of Computational Methods in Engineering, 29: 3567–3586.
Mokhtar S, Sojka A, Davila CC (2020). Conditional generative adversarial networks for pedestrian wind flow approximation. In: Proceedings of the 11th Annual Symposium on Simulation for Architecture and Urban Design.
O’Shea K, Nash R (2015). An introduction to convolutional neural networks. arXiv: 1511.08458.
Pérez-Sánchez B, Fontenla-Romero O, Guijarro-Berdiñas B (2018). A review of adaptive online learning for artificial neural networks. Artificial Intelligence Review, 49: 281–299.
Pope SB (1997). Computationally efficient implementation of combustion chemistry using in situ adaptive tabulation. Combustion Theory and Modelling, 1: 41–63.
Ramanan N, Homsy GM (1994). Linear stability of lid-driven cavity flow. Physics of Fluids, 6: 2690–2701.
Smith LN (2018). A disciplined approach to neural network hyperparameters: part 1—learning rate, batch size, momentum, and weight decay. arXiv: 1803.09820.
Sun L, Chen J, Xu Y, et al. (2022). Hierarchical amortized GAN for 3D high resolution medical image synthesis. IEEE Journal of Biomedical and Health Informatics, 26: 3966–3975.
Tian W, Sevilla TA, Li D, et al. (2018). Fast and self-learning indoor airflow simulation based on in situ adaptive tabulation. Journal of Building Performance Simulation, 11: 99–112.
Wang L, Chen Q (2008). Applications of a coupled multizone-CFD model to calculate airflow and contaminant dispersion in built environments for emergency management. HVAC&R Research, 14: 925–939.
Wang M, Chen Q (2009). Assessment of various turbulence models for transitional flows in an enclosed environment (RP-1271). HVAC&R Research, 15: 1099–1119.
Wang M, Chen Q (2010). On a hybrid RANS/LES approach for indoor airflow modeling (RP-1271). HVAC&R Research, 16: 731–747.
Wang Y, Wang W, Tao G, et al. (2022). Optimization of the semi-sphere vortex generator for film cooling using generative adversarial network. International Journal of Heat and Mass Transfer, 183: 122026.
Ye Y, Strong M, Lou Y, et al. (2022). Evaluating performance of different generative adversarial networks for large-scale building power demand prediction. Energy and Buildings, 269: 112247.
Zeiler MD, Krishnan D, Taylor GW, et al. (2010). Deconvolutional networks. In: Proceedings of 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, San Francisco, CA, USA.
Zhang Q, Ferdowsi A, Saad W, et al. (2022). Distributed conditional generative adversarial networks (GANs) for data-driven millimeter wave communications in UAV networks. IEEE Transactions on Wireless Communications, 21: 1438–1452.
Zhao X, Ma F, Güera D, et al. (2022). Generative multiplane images: Making a 2D GAN 3D-Aware. In: Proceedings of European Conference on Computer Vision.
Zheng Y, Zhang Y, Zheng Z (2021). Continuous conditional generative adversarial networks (cGAN) with generator regularization. arXiv: 2103.14884.
Zhou Q, Ooka R (2020). Comparison of different deep neural network architectures for isothermal indoor airflow prediction. Building Simulation, 13: 1409–1423.
Zhu X, Shi T, Jin X, et al. (2021). Multi-sensor information fusion based control for VAV systems using thermal comfort constraints. Building Simulation, 14: 1047–1062.
Zuo W, Chen Q (2009). Real-time or faster-than-real-time simulation of airflow in buildings. Indoor Air, 19: 33–44.
Zuo W, Chen Q (2010). Fast and informative flow simulations in a building by using fast fluid dynamics model on graphics processing unit. Building and Environment, 45: 747–757.
Acknowledgements
This research was supported in part by the U.S. Defense Threat Reduction Agency and performed under U.S. Department of Energy Contract No. DE-AC02-05CH11231. This research was also partially supported by the National Science Foundation under Awards No. IIS-1802017, CBET-2217410, CNS-2025377, and CNS-2241361.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Faulkner, C.A., Jankowski, D.S., Castellini, J.E. et al. Fast prediction of indoor airflow distribution inspired by synthetic image generation artificial intelligence. Build. Simul. 16, 1219–1238 (2023). https://doi.org/10.1007/s12273-023-0989-1
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12273-023-0989-1