Reinforcement learning is applied to the development of control strategies in order to reduce skin friction drag in a fully developed turbulent channel flow at a low Reynolds number. Motivated by the so-called opposition control (Choi et al., J. Fluid Mech., vol. 253, 1993, pp. 509–543), in which a control input is applied so as to cancel the wall-normal velocity fluctuation on a detection plane at a certain distance from the wall, we consider wall blowing and suction as a control input, and its spatial distribution is determined by the instantaneous streamwise and wall-normal velocity fluctuations at distance 15 wall units above the wall. A deep neural network is used to express the nonlinear relationship between the sensing information and the control input, and it is trained so as to maximize the expected long-term reward, i.e. drag reduction. When only the wall-normal velocity fluctuation is measured and a linear network is used, the present framework reproduces successfully the optimal linear weight for the opposition control reported in a previous study (Chung & Talha, Phys. Fluids, vol. 23, 2011, 025102). In contrast, when a nonlinear network is used, more complex control strategies based on the instantaneous streamwise and wall-normal velocity fluctuations are obtained. Specifically, the obtained control strategies switch abruptly between strong wall blowing and suction for downwelling of a high-speed fluid towards the wall and upwelling of a low-speed fluid away from the wall, respectively. Extracting key features from the obtained policies allows us to develop novel control strategies leading to drag reduction rates as high as 37 %, which is higher than the 23 % achieved by the conventional opposition control at the same Reynolds number. Finding such an effective and nonlinear control policy is quite difficult by relying solely on human insights. The present results indicate that reinforcement learning can be a novel framework for the development of effective control strategies through systematic learning based on a large number of trials.

Sustained friction drag reduction and heat transfer augmentation are simultaneously achieved in a fully developed channel flow where the averaged transport equations and wall boundary conditions for momentum and heat have identical form. Zero-net-mass-flux wall blowing and suction is assumed as a control input and its spatio-temporal distribution is determined based on optimal control theory. When the root-mean-square value of the control input is 5 % of the bulk mean velocity, the friction drag is decreased by 24 % from the uncontrolled value, whereas the heat transfer is more than doubled. Optimizations with different amplitudes of the control input and different Reynolds numbers reveal that the optimal control inputs commonly exhibit the property of a downstream travelling wave, whose wavelength is ∼250 in wall units and phase velocity is ∼30 % of the bulk mean velocity. Detailed analyses of the controlled velocity and thermal fields show that the travelling wave input contributes to dissimilar heat transfer enhancement through two distinct mechanisms, i.e. direct modification of the coherent velocity and thermal fields and an indirect effect on the random fields. The present results show that the divergence-free velocity vector and the conservative scalar are essentially different, and this is a key to achieving dissimilar heat transfer enhancement in turbulent shear flows.

High-precisiondrop-on-demand (DOD) inkjet printing has been considered as one of promising technologies for the fabrication of advanced functional materials. For a DOD printer, high-precision dispensing techniques for achieving satellite-free smaller droplets, have long been desired for patterning thin-film structures. Optimization of actuation waveform driving a DOD inkjet printer is one of the most versatile and effective strategies to obtain high-precision droplets. Considering the complexity of physics behind the droplet dispensing mechanisms and the large degrees of freedom in the applied waveforms, conventional trial-and-error approaches are not effective for searching the optimal waveform. The present study considers the inlet velocity of a liquid chamber located upstream of a dispensing nozzle as a control variable and aims to develop an automated waveform tunning framework to optimize its waveform using a sample-efficient Bayesian optimization algorithm. Firstly, the droplet dispensing dynamics are numerically reproduced by using an open-source OpenFOAM solver, interFoam, and the results are passed on to another code based on pyFoam. Then, the parameters characterizing the actuation waveform driving a DOD printer are determined by the Bayesian optimization (BO) algorithm so as to maximize a prescribed multi-objective function expressed as the sum of two factors, i.e., the size of a primary droplet and the presence of satellite droplets. The results show that the present BO algorithm can successfully find high-precision dispensing waveforms within 150 simulations. Specifically, satellite droplets can be effectively eliminated and the droplet diameter can be significantly reduced to 24.9% of the nozzle diameter by applying the optimal waveform.

A series of direct numerical simulations of a fully developed turbulent channel flow is conducted in order to clarify the effects of travelling wave-like wall blowing and suction on dissimilar heat transfer enhancement. While the wave form is kept sinusoidal and its amplitude is set to be 5 % of the bulk mean velocity, the wavelength and phase speed of the travelling wave are systematically changed in a wide parameter space. As a result, the global optimum of the parameter set for maximizing the analogy factor, which is defined as the ratio between the Stanton number and the skin-friction coefficient, is identified. Interestingly, the obtained globally optimal mode agrees well with that predicted from the optimal control theory taking into account the future dynamics within a limited time horizon by Yamamoto et al. (J. Fluid Mech., vol. 733, 2013, pp. 189–220). The instantaneous velocity and thermal fields are decomposed into coherent and random components in order to evaluate the contribution from each component to dissimilar heat transfer enhancement. The detailed mechanisms of dissimilarity are explained by the budget analyses of the coherent and random contributions. Also, their relationships with the near-wall turbulent structures modified by the applied control are discussed through flow visualization. It is found that the random component makes a dominant contribution to dissimilarity, and this can be explained by an indirect effect through the modification of the coherent field by the applied control. Based on the above mechanisms, we propose a simple unsteady Reynolds-averaged Navier–Stokes (URANS) approach, where the phase-averaged velocity and thermal fields are solved directly whereas the effects of the random component are modelled by the Boussinesq eddy viscosity and diffusivity hypothesis. It is shown that the present URANS can capture the overall trend of dissimilar heat transfer enhancement in a wide parameter range. The present results also explain why the optimal control theory with a limited time horizon succeeds in predicting the globally optimal control mode.