Core position alignment in polymer optical waveguides fabricated using the Mosquito method.

We fabricate multimode polymer optical waveguides with circular graded-index (GI) cores which are aligned in parallel at desired positions using the Mosquito method. In the Mosquito method, three-dimensional wiring patterns can be formed with a simple process. However, the core position is likely to deviate from the designed position because of multiple fabrication factors. Hence, in this paper, the dominant parameters to influence on the core height in the cladding are investigated both theoretically and experimentally. In particular, a linear relationship between the core height and the needle-tip height is confirmed with theoretical fluid analysis. Using this relationship, we succeeded in fabricating a waveguide in which the maximum variation of the core height from the designed value is controlled to be less than 10 µm.

Identifying key differences between linear stochastic estimation and neural networks for fluid flow regressions.

Neural networks (NNs) and linear stochastic estimation (LSE) have widely been utilized as powerful tools for fluid-flow regressions. We investigate fundamental differences between them considering two canonical fluid-flow problems: (1) the estimation of high-order proper orthogonal decomposition coefficients from low-order their counterparts for a flow around a two-dimensional cylinder, and (2) the state estimation from wall characteristics in a turbulent channel flow. In the first problem, we compare the performance of LSE to that of a multi-layer perceptron (MLP). With the channel flow example, we capitalize on a convolutional neural network (CNN) as a nonlinear model which can handle high-dimensional fluid flows. For both cases, the nonlinear NNs outperform the linear methods thanks to nonlinear activation functions. We also perform error-curve analyses regarding the estimation error and the response of weights inside models. Our analysis visualizes the robustness against noisy perturbation on the error-curve domain while revealing the fundamental difference of the covered tools for fluid-flow regressions.

Control of flow around a circular cylinder for minimizing energy dissipation.

Control of flow around a circular cylinder is studied numerically aiming at minimization of the energy dissipation. First, we derive a mathematical relationship (i.e., identity) between the energy dissipation in an infinitely large volume and the surface quantities, so that the cost function can be expressed by the surface quantities only. Subsequently a control law to minimize the energy dissipation is derived by using the suboptimal control procedure [J. Fluid Mech. 401, 123 (1999)JFLSA70022-112010.1017/S002211209900659X]. The performance of the present suboptimal control law is evaluated by a parametric study by varying the value of the arbitrary parameter contained. Two Reynolds numbers, Re=100 and 1000, are investigated by two-dimensional simulations. Although no improvement is obtained at Re=100, the present suboptimal control shows better results at Re=1000 than the suboptimal controls previously proposed. With the present suboptimal control, the dissipation and the drag are reduced by 58% and 44% as compared to the uncontrolled case, respectively. The suction around the front stagnation point and the blowing in the rear half are found to be weakened as compared to those in the previous suboptimal control targeting at pressure drag reduction. A predetermined control based on the control input profile obtained by the suboptimal control is also performed. The energy dissipation and the drag are found to be reduced as much as those in the present suboptimal control. It is also found that the present suboptimal and predetermined controls have better energy efficiencies than the suboptimal control previously proposed. Investigation at different control amplitudes reveals an advantage of the present control at higher amplitude. Toward its practical implementation, a localized version of the predetermined control is also examined, and it is found to work as effectively as the continuous case. Finally, the present predetermined control is confirmed to work well in a three-dimensional flow too.

Active control of vortex shedding: an explanation of the gain window.

This paper explains the gain window phenomenon seen in early experimental and computational studies on active, closed-loop control of vortex shedding, whereby shedding is completely suppressed only if the feedback gain lies within some narrow window of stabilizing gains. Using two-dimensional direct numerical simulations and reduced-order modeling techniques, a low-order, linear model of the cylinder wake is formed at a Reynolds number of 60. This model is used to reproduce and to explain the gain window seen in previous studies. It is shown that the gain window is not caused by the destabilization of a higher mode but rather is determined entirely by the behavior of the open-loop unstable mode under the action of the closed-loop controller. It is demonstrated that the time taken for actuated fluid to convect to the sensor location plays an important part in explaining this gain window. A similar analysis at a higher Reynolds number of 80 reveals that the wake remains unstable for all choices of the feedback gain. The study illustrates the limitations of closed-loop suppression of vortex shedding when a very simple control strategy is used.

Phase relationship in laminar channel flow controlled by traveling-wave-like blowing or suction.

The phase relationship between the streamwise and the wall-normal velocity disturbances induced by a traveling-wave-like blowing or suction control [T. Min, J. Fluid Mech. 558, 309 (2006)] in a two-dimensional laminar Poiseuille flow is investigated. The investigation is done by solving the linearized Navier-Stokes equation and by using the identity equation between the skin-friction drag and the Reynolds shear stress [K. Fukagata, Phys. Fluids 14, L73 (2002)]. It has been known that a traveling wave creates a nonquadrature between the velocity disturbances and generates the positive phase shift of the streamwise velocity disturbance in the case of a skin-friction drag reduction. The present analysis further reveals that this nonquadrature consists of an inviscid base phase relationship and a near-wall phase shift induced by the viscosity. The analogy between the present control and Stokes' second problem is discussed. The thickness of the near-wall region in which the viscous phase shift takes place is found to be scaled similarly to the Stokes' second problem.