Abstract
In this work, we develop a method for integrating an AI model with a CFD solver to predict chemical mixing simulations’ output. The proposed AI model is based on a deep neural network with a variational autoencoder that is managed by our AI supervisor. We demonstrate that the developed method allows us to accurately accelerate the steady-state simulations of chemical reactions performed with the MixIT solver from Tridiagonal solutions.
In this paper, we investigate the accuracy and performance of AI-accelerated simulations, considering three different scenarios: i) prediction in cases with the same geometry of mesh as used during training the model, ii) with a modified geometry of tube in which the ingredients are mixed, iii) with a modified geometry of impeller used to mix the ingredients.
Our AI model is trained on a dataset containing 1500 samples of simulated scenarios and can accurately predict the process of chemical mixing under various conditions. We demonstrate that the proposed method achieves accuracy exceeding 90% and reduces the execution time up to 9 times.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Huang, K., Krügener, M., Brown, A., Menhorn, F., Bungartz, H.J., Hartmann, D.: Machine Learning-Based Optimal Mesh Generation in Computational Fluid Dynamics. ArXiv, pp. 1–22 (2021). https://arxiv.org/abs/2102.12923
Iserte, S., Macías, A., Martínez-Cuenca, R., Chiva, S., Paredes, R., Quintana-Ortí, E.S.: Accelerating urban scale simulations leveraging local spatial 3D structure. J. Comput. Sci. 62, 101741 (2022)
Kim, B., Azevedo, V.C., Thuerey, N., Kim, T., Gross, M., Solenthaler, B.: Deep fluids: a generative network for parameterized fluid simulations. Comput. Graph. Forum 38(2), 59–70 (2019)
Kochkov, D., Smith, J.A., Alieva, A., Wang, Q., Brenner, M.P., Hoyer, S.: Machine learning-accelerated computational fluid dynamics. Proceed. Nat. Acad. Sci. 118(21), e2101784118 (2021). https://doi.org/10.1073/pnas.2101784118. https://www.pnas.org/doi/abs/10.1073/pnas.2101784118
Kurz, M., Offenhäuser, P., Viola, D., Shcherbakov, O., Resch, M., Beck, A.: Deep reinforcement learning for computational fluid dynamics on HPC systems. J. Comput. Sci. 65, 101884 (2022). https://doi.org/10.1016/j.jocs.2022.101884. https://www.sciencedirect.com/science/article/pii/S1877750322002435
MixIT: the enterprise mixing analysis tool. https://mixing-solution.com/. Accessed 19 Apr 2023
Obiols-Sales, O., Vishnu, A., Malaya, N., Chandramowliswharan, A.: CFDNet: a deep learning-based accelerator for fluid simulations. Proceed. Int. Conf. Supercomput. (2020). https://doi.org/10.1145/3392717.3392772
One-Class Support Vector Machine (SVM) for Anomaly Detection. https://grabngoinfo.com/one-class-support-vector-machine-svm-for-anomalydetection/. Accessed 26 Sept 2021
OpenFOAM. http://www.openfoam.com/. Accessed 19 Apr 2023
OpenFOAM: User Guide: k-epsilon. http://www.openfoam.com/%20documentation/guides/latest/doc/guide-turbulence-ras-k-epsilon.html. Accessed 28 Feb 2023
Panchigar, D., Kar, K., Shukla, S., Mathew, R.M., Chadha, U., Selvaraj, S.K.: Machine learning-based CFD simulations: a review, models, open threats, and future tactics. Neural Comput. Appl. 34(24), 21677–21700 (2022). https://doi.org/10.1007/s00521-022-07838-6
Roh, S., Song, H.J.: Evaluation of Neural Network Emulations for Radiation Parameterization in Cloud Resolving Model. Geophys. Res. Lett. 47(21), e2020GL089444 (2020)
Rojek, K.: Machine learning method for energy reduction by utilizing dynamic mixed precision on GPU-based supercomputers. Concurr. Comput. Practice Exper. 31(6), e4644 (2019). https://doi.org/10.1002/cpe.4644. https://onlinelibrary.wiley.com/doi/abs/10.1002/cpe.4644
Rojek, K., Wyrzykowski, R.: Performance and scalability analysis of AI-accelerated CFD simulations across various computing platforms. In: Singer, J., Elkhatib, Y., Blanco Heras, D., Diehl, P., Brown, N., Ilic, A. (eds.) Euro-Par 2022: Parallel Processing Workshops. Euro-Par 2022. Lecture Notes in Computer Science, vol. 13835. Springer, Cham (2023). https://doi.org/10.1007/978-3-031-31209-0_17
Rojek, K., Wyrzykowski, R., Gepner, P.: AI-accelerated CFD simulation based on OpenFOAM and CPU/GPU computing. In: Paszynski, M., Kranzlmüller, D., Krzhizhanovskaya, V.V., Dongarra, J.J., Sloot, P.M.A. (eds.) ICCS 2021. LNCS, vol. 12743, pp. 373–385. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-77964-1_29
Savarese, M., Cuoci, A., De Paepe, W., Parente, A.: Machine learning clustering algorithms for the automatic generation of chemical reactor networks from CFD simulations. Fuel 343, 127945 (2023). https://doi.org/10.1016/j.fuel.2023.127945. https://www.sciencedirect.com/science/article/pii/S0016236123005586
Schannen, M., Bachem, O., Lucic, M.: Recent advances in autoencoder-based representation learning. arXiv preprint arXiv:1812.05069 (2018)
Song, H.J., Kim, P.S.: Effects of Cloud Microphysics on the Universal Performance of Neural Network Radiation Scheme. Geophys. Res. Lett. 49(9), e2022GL098601 (2022)
Song, H.J., et al.: Benefits of stochastic weight averaging in developing neural network radiation scheme for numerical weather prediction. J. Adv. Model. Earth Syst. 14(10), e2021MS002921 (2022)
Srivastava, S., Damodaran, M., Khoo, B.C.: Machine learning surrogates for predicting response of an aero-structural-sloshing system. arXiv preprint (2019)
Statistics How To. https://www.statisticshowto.com/probability-and-statistics/calculus-based-statistics/. Accessed 28 Feb 2023
TensorFlow Core: Mixed precision. http://www.tensorflow.org/guide/mixed_%20precision. Accessed 19 Apr 2023
Tompson, J., Schlachter, K., Sprechmann, P., Perlin, K.: Accelerating Eulerian fluid simulation with convolutional networks. In: Proceedings 34th International Conference Machine Learning, ICML2017 - Vol. 70, pp. 3424–3433 (2017)
Vinuesa, R., Brunton, S.L.: The potential of machine learning to enhance computational fluid dynamics. arXiv preprint arXiv:2110.02085 (2021)
Wyatt, M.R., Yamamoto, V., Tosi, Z., Karlin, I., Essen, B.V.: Is Disaggregation possible for HPC Cognitive Simulation? In: 2021 IEEE/ACM Workshop on Machine Learning in High Performance Computing Environments (MLHPC), pp. 94–105 (2021). https://doi.org/10.1109/MLHPC54614.2021.00014
Xiao, D., et al.: A reduced order model for turbulent flows in the urban environment using machine learning. Build. Environ. 148, 323–337 (2019)
Yu, W., Zhao, F., Yang, W., Xu, H.: Integrated analysis of CFD simulation data with K-means clustering algorithm for soot formation under varied combustion conditions. Appl. Thermal Eng. 153, 299–305 (2019). https://doi.org/10.1016/j.applthermaleng.2019.03.011. https://www.sciencedirect.com/science/article/pii/S1359431118349172
Acknowledgements
This research was supported by project no. 020/RID/2018 /19 financed within the program of the Polish Ministry of Science and Higher Education “Regional Initiative of Excellence” (years 2019 - 2023, the amount of financing 12 000 000 PLN). This research was partly supported by the PLGrid infrastructure at ACK Cyfronet AGH in Krakow.
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 Switzerland AG
About this paper
Cite this paper
Rojek, K., Wyrzykowski, R., Gepner, P. (2023). Chemical Mixing Simulations with Integrated AI Accelerator. In: Mikyška, J., de Mulatier, C., Paszynski, M., Krzhizhanovskaya, V.V., Dongarra, J.J., Sloot, P.M. (eds) Computational Science – ICCS 2023. ICCS 2023. Lecture Notes in Computer Science, vol 14074. Springer, Cham. https://doi.org/10.1007/978-3-031-36021-3_50
Download citation
DOI: https://doi.org/10.1007/978-3-031-36021-3_50
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-36020-6
Online ISBN: 978-3-031-36021-3
eBook Packages: Computer ScienceComputer Science (R0)