17 March 2025 in Physical Review Fluids
6 February 2026 in Journal of Fluid Mechanics
The proposed study aims to optimise a real-time opposition control strategy to reduce the intensity of near-wall sweep events by applying a Bayesian optimisation algorithm. The experiments were conducted in a fully turbulent channel flow characterised by a friction Reynolds number of $350$ . Sweep events were identified using a gradient-based detection technique and controlled via a wall-normal jet. An open-loop control logic was implemented and the control parameters (frequency, voltage amplitude and delay time) were optimised, within the bounds imposed by the experimental set-up, to bring the maximum sweep events intensity reduction up to $54\,\%$ , with a robust cost function. The effects of the control were observed by analysing the conditionally averaged sweep events at various streamwise locations downstream of the actuation point. Moreover, the conditional analysis was applied to the cross-correlation function of velocity signals highlighting the large reduction of the sweep event convection velocity during the blowing phase of the jet. An overall energy increase has been found in the conditionally averaged energy spectra for the controlled case. The analysis of conditionally averaged wavelet spectra revealed that the control, by interrupting the natural evolution of the sweep event, initially leads to a reduction in the energy associated with it, followed by a subsequent increase during the development of the jet-blowing phase.
1 August 2026 in Physics of Fluids
This study investigates the use of a feed-forward neural network (FNN) and a hybrid approach combining an FNN with proper orthogonal decomposition (POD) to estimate the effect of control on sweep events within an opposition control framework. Conditionally averaged sweep events (CASEs) were experimentally sampled in a fully turbulent channel flow at a friction Reynolds number of 350. The two architectures were then trained to reconstruct the controlled CASEs using the control parameters (frequency, voltage amplitude, and delay time) as input. The results indicate that both approaches are able to reproduce the control signature across different actuation parameters with comparable accuracy. At the same time, the hybrid POD–FNN architecture requires over two orders of magnitude fewer trainable parameters. Finally, both local and global sensitivity analyses were performed, leveraging the differentiability of the trained models and their rapid inference times. The Jacobian matrix and Sobol indices were computed to conduct a local and a global sensitivity analysis, respectively. The findings revealed a high sensitivity to the actuation frequency and delay time parameters, whereas a lower sensitivity was observed with respect to the voltage amplitude.