Abstract
A key question in flow control is that of the design of optimal controllers when the control space is high-dimensional and the experimental or computational budget is limited. We address this formidable challenge using a particular flavor of machine learning and present the first application of Bayesian optimization to the design of open-loop controllers for fluid flows. We consider a range of acquisition functions, including the recently introduced output-informed criteria of Blanchard and Sapsis (2021), and evaluate performance of the Bayesian algorithm in two iconic configurations for active flow control: computationally, with drag reduction in the fluidic pinball; and experimentally, with mixing enhancement in a turbulent jet. For these flows, we find that Bayesian optimization identifies optimal controllers at a fraction of the cost of other optimization strategies considered in previous studies. Bayesian optimization also provides, as a by-product of the optimization, a surrogate model for the latent cost function, which can be leveraged to paint a complete picture of the control landscape. The proposed methodology can be used to design open-loop controllers for virtually any complex flow and, therefore, has significant implications for active flow control at an industrial scale.
Graphic Abstract
Similar content being viewed by others
References
Gad-el Hak, M.: Flow Control: Passive, Active, and Reactive Flow Management. Cambridge University Press, Cambridge (2007)
Brunton, S.L., Noack, B.R.: Closed-loop turbulence control: progress and challenges. Appl. Mech. Rev. 67(5), 050801 (2015)
Gad-el Hak, M.: Modern developments in flow control. Appl. Mech. Rev. 49, 365 (1996)
Bewley, T.R.: Flow control: new challenges for a new renaissance. Prog. Aerosp. Sci. 37, 21 (2001)
Cattafesta, L.N., III., Sheplak, M.: Actuators for active flow control. Annu.Rev. Fluid Mech. 43, 247 (2011)
Choi, H., Jeon, W.P., Kim, J.: Actuators for active flow control. Annu. Rev. Fluid Mech. 40, 113 (2008)
Pastoor, M., Henning, L., Noack, B.R., et al.: Feedback shear layer control for bluff body drag reduction. J. Fluid Mech. 608, 161 (2008)
Aamo, O.M., Krstic, M.: Flow Control by Feedback: Stabilization and Mixing. Springer, London (2003)
Dimotakis, P.E.: Turbulent mixing. Annu. Rev. Fluid Mech. 37, 329 (2005)
Dowling, A.P., Morgans, A.S.: Feedback control of combustion oscillations. Annu. Rev. Fluid Mech. 37, 151 (2005)
Bagheri, S., Henningson, D.S.: Transition delay using control theory. Philos. Trans. R. Soc. A 369, 1365 (2011)
Fabbiane, N., Semeraro, O., Bagheri, S., et al.: Adaptive and model-based control theory applied to convectively unstable flows. Appl. Mech. Rev. 66(6), 060801 (2014)
Rowley, C.W., Dawson, S.T.M.: Model reduction for flow analysis and control. Annu. Rev. Fluid Mech. 49, 387 (2017)
Li, Y., Cui, W., Jia, Q., et al.: Explorative gradient method for active drag reduction of the fluidic pinball and slanted Ahmed body. J. Fluid Mech. (2021). arXiv:1905.12036
Jones, D.R., Schonlau, M., Welch, W.J.: Efficient global optimization of expensive black-box functions. J. Glob. Optim. 13, 455 (1998)
Hennig, P., Schuler, C.J.: Entropy search for information-efficient global optimization. J. Mach. Learn. Res. 13, 1809 (2012)
Duriez, T., Brunton, S.L., Noack, B.R.: Machine Learning Control—Taming Nonlinear Dynamics and Turbulence. Springer, Cham (2017)
Kutz, J.N.: Deep learning in fluid dynamics. J. Fluid Mech. 814, 1 (2017)
Brunton, S.L., Noack, B.R., Koumoutsakos, P.: Machine learning for fluid mechanics. Annu. Rev. Fluid Mech. 52, 477 (2020)
Fernex, D., Semaan, R., Albers, M., Meysonnat, P.S., Schröder, W., Noack, B.R.: Actuation response model from sparse data for wall turbulence drag reduction. Phys. Rev. Fluids 5, 073901 (2020)
Lee, C., Kim, J., Babcock, D., Goodman, R.: Application of neural networks to turbulence control for drag reduction. Phys. Fluids 9(6), 1740 (1997)
Efe, M., Debiasi, M., Yan, P., Ozbay, H., Samimy, M. :Control of subsonic cavity flows by neural networks-analytical models and experimental validation. In: 43rd AIAA Aerospace Sciences Meeting and Exhibit, p. 294 (2005)
Park, J., Choi, H.: Machine-learning-based feedback control for drag reduction in a turbulent channel flow. J. Fluid Mech. 904, A24 (2020)
Benard, N., Pons-Prats, J., Periaux, J., et al.: Turbulent separated shear flow control by surface plasma actuator: experimental optimization by genetic algorithm approach. Exp. Fluids 57, 22 (2016)
Ren, F., Hu, H.B., Tang, H.: Active flow control using machine learning: a brief review. J. Hydrodyn. 32, 247 (2020)
Zhou, Y., Fan, D., Zhang, B., et al.: Artificial intelligence control of a turbulent jet. J. Fluid Mech. 897, A27:1 (2020)
Rabault, J., Kuchta, M., Jensen, A., et al.: Artificial neural networks trained through deep reinforcement learning discover control strategies for active flow control. J. Fluid Mech. 865, 281 (2019)
Brochu, E., Cora, V.M., De Freitas, N.: A tutorial on Bayesian optimization of expensive cost functions, with application to active user modeling and hierarchical reinforcement learning. arXiv preprint (2010). arXiv:1012.2599
Shahriari, B., Swersky, K., Wang, Z., et al.: Taking the human out of the loop: a review of Bayesian optimization. Proc. IEEE 104, 148 (2015)
Frazier, P.I.: Bayesian optimization. In: Recent Advances in Optimization and Modeling of Contemporary Problems (INFORMS, 2018), pp. 255–278 (2018)
Blanchard, A., Sapsis, T.: Bayesian optimization with output-weighted optimal sampling. J. Comput. Phys. 425, 109901 (2021)
Deng, N., Noack, B.R., Morzyński, M., Pastur, R.: Low-order model for successive bifurcations of the fluidic pinball. J. Fluid Mech. 884, A37 (2020)
Chen, W., Ji, C., Alam, M.M., Williams, J., Xu, D.: Numerical simulations of flow past three circular cylinders in equilateral-triangular arrangements. J. Fluid Mech. 891, A14 (2020)
Cornejo Maceda, G.Y., Li, Y., Lusseyran, F., et al.: Stabilization of the fluidic pinball with gradient-enriched machine learning control. J. Fluid Mech. 917, A42 (2021)
Koumoutsakos, P., Freund, J., Parekh, D.: Evolution strategies for automatic optimization of jet mixing. AIAA J. 39, 967 (2001)
Rasmussen, C.E., Williams, C.K.I.: Gaussian Processes for Machine Learning. MIT, Cambridge (2006)
Raissi, M., Perdikaris, P., et al.: Machine learning of linear differential equations using Gaussian processes. J. Comput. Phys. 348, 683 (2017)
Pang, G., Perdikaris, P., Cai, W., Karniadakis, G.E.: Discovering variable fractional orders of advection-dispersion equations from field data using multi-fidelity Bayesian optimization. J. Comput. Phys. 348, 694 (2017)
Srinivas, N., Krause, A., Kakade, S.M., et al.: Gaussian process optimization in the bandit setting: No regret and experimental design. arXiv preprint (2009). arXiv:0912.3995
Sacks, J., Welch, W.J., Mitchell, T.J., et al.: Design and analysis of computer experiments. Stat. Sci. 4, 409–423 (1989)
Yang, Y., Blanchard, A., Sapsis, T., et al.: Output-weighted sampling for multi-armed bandits with extreme payoffs. arXiv preprint (2021). arXiv:2102.10085
Blanchard, A., Sapsis, T.: Output-weighted optimal sampling for Bayesian experimental design and uncertainty quantification. SIAM/ASA J. Uncertainty Quant. 9, 564 (2021)
Fischer, P.F., Lottes, J.W. , Kerkemeier, S.G.: Nek5000 web page (2008). http://nek5000.mcs.anl.gov
Raibaudo, C., Zhong, P., Noack, B.R., et al.: Machine learning strategies applied to the control of a fluidic pinball. Phys. Fluids 32, 015108 (2020)
Perumal, A.K., Zhou, Y.: Parametric study and scaling of jet manipulation using an unsteady minijet. J. Fluid Mech. 848, 592 (2018)
Nair, A.G., Yeh, C.A., Kaiser, E., et al.: Cluster-based feedback control of turbulent post-stall separated flows. J. Fluid Mech. 875, 345 (2019)
Cornejo Maceda, G.Y.: Gradient-enriched machine learning control exemplified for shear flows in simulations and experiments. Université Paris-Saclay (2021). [Ph.D. thesis]
Acknowledgements
ABB and TPS gratefully acknowledge support from a MathWorks Faculty Research Innovation Fellowship at MIT. GYCM and BRN acknowledge funding by the French National Research Agency (ANR) via the grant FlowCon (ANR-17-ASTR-0022). The thesis of GYCM was supported by LIMSI/CNRS and Paris-Sud University. BRN acknowledges support from the National Science Foundation of China (NSFC) through grant 12172109. YZ wishes to acknowledge support given to him from NSFC through grants 11632006, 91752109 and 91952204.
Author information
Authors and Affiliations
Corresponding authors
Additional information
Executive Editor: Weiwei Zhang.
Appendix A: Validation of the computational approach for the fluidic pinball
Appendix A: Validation of the computational approach for the fluidic pinball
To determine appropriate values for the time-step size \(\Delta \tau \) and polynomial order N, we consider the uncontrolled case (\(\mathbf{b } = \mathbf{0 }\)) at \(Re=100\) with initial condition \(\mathbf{v }(x, y) = [1+ 10^{-3} \sin (y)] \mathbf{e }_x\). The slight asymmetry in the initial condition allows vortex shedding to develop more rapidly than if one were to rely on small asymmetries in the numerics. The flow is evolved for 2000 time units and only the last 500 time units of that interval are retained to compute the statistics. We report the temporal mean, standard deviation (std), and peak-to-peak amplitude (amp) of \(C_D\) and \(C_L\), as well as the Strouhal frequency St (i.e., the dominant frequency of \(C_L\)). Table 2 shows that the polynomial order has a larger effect on the statistics than the time-step size. None of the computational parameters has a significant effect on the Strouhal frequency. The results show that, for the present purposes, adequate convergence is achieved by specifying \(\Delta \tau =0.02\) and \(N=7\). In our production runs, however, we use a smaller time-step size (\(\Delta \tau =0.01\)) to alleviate the possibility of solution blow-up when the actuation vector is non-zero.
For \(\Delta \tau =0.01\) and \(N=7\), we compare our computational approach with the finite-element solver used by Deng et al. [32]. We consider the actuation configuration described in Cornejo Maceda et al. [34] in which the front cylinder is not allowed to rotate (\(v_1=0\)) and the aft cylinders rotate with equal and opposite angular velocities (\(v_2 = -v_3\)). As in Cornejo Maceda et al. [34], the Reynolds number is 100 and the flow is effected for 1000 time units starting from an initial state on the limit cycle of the uncontrolled configuration. We report the temporal mean for \(C_D\) and \(J_T\). Figure 12 shows good agreement between our approach and that of Cornejo Maceda et al. [34, 47].
Rights and permissions
About this article
Cite this article
Blanchard, A.B., Cornejo Maceda, G.Y., Fan, D. et al. Bayesian optimization for active flow control. Acta Mech. Sin. 37, 1786–1798 (2021). https://doi.org/10.1007/s10409-021-01149-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10409-021-01149-0