Non-intrusive reduced genetic algorithm for near-real time flow optimal control

Most genetic algorithms (GAs) used in the literature to solve control problems are time consuming and involve important storage memory requirements. In fact, the search in GAs is iteratively performed on a population of chromosomes (control parameters). As a result, the cost functional needs to be evaluated through solving the high fidelity model or by performing the experimental protocol for each chromosome and for many generations. To overcome this issue, a non-intrusive reduced real-coded genetic algorithm (RGA) for near real-time optimal control is designed. This algorithm uses precalculated parametrized solution snapshots stored in the POD (proper orthogonal decomposition) reduced form, to predict the solution snapshots for chromosomes over generations. The method used for this purpose is a economic reduced version of the Bi-CITSGM method (Bi-calibrated interpolation on the tangent space of the Grassmann manifold) designed specially for nonlinear parametrized solution snapshots interpolation. This approach is proposed in such a way to accelerate the usual Bi-CITSGM process by bringing this last to a significantly lower dimension. Thus, the whole optimization process by RGA can be performed in near real-time. The potential of RGA in terms of accuracy and central processing unit time is demonstrated on control problems of the flow past a cylinder and flow in a lid-driven cavity when the Reynolds number value varies.     

Experimental and computational study of the flow induced by a plasma actuator

A complementary experimental and computational study of the flow field evoked by a plasma actuator mounted on a flat plate was in focus of the present work. The main objective of the experimental investigation was the determination of the vector force imparted by the plasma actuator to the fluid flow. The force distribution was presently extracted from the Navier–Stokes equations directly by feeding them with the velocity field measured by a PIV technique. Assuming a steady-in-mean, two-dimensional flow with zero-pressure gradient, the imbalance between the convective term and the momentum equation’s right-hand-side terms reveals the desired resulting force. This force-distribution database was used afterwards as the source term in the momentum equation. Furthermore, an empirical model formulation for the volume-force determination parameterized by the underlying PIV-based model is derived. The model is tested within the RANS framework in order to predict a wall jet-like flow induced by a plasma actuator. The Reynolds equations are closed by a near-wall second-moment closure model based on the homogeneous dissipation rate of the kinetic energy of turbulence. The computationally obtained velocity field is analysed along with the experimental data focussing on the wall jet flow region in proximity of the plasma actuator. For comparison purposes, different existing phenomenological models were applied to evaluate the new model’s accuracy. The comparative analysis of all applied models demonstrates the strength of the new empirical model, particularly within the plasma domain. In addition, the presently formulated empirical model was applied to the flow in a three-dimensional diffuser whose inflow was modulated by a pair of streamwise vortices generated by the present plasma actuator. The direct comparison with existing experimental data of Grundmann et al. (2011) demonstrated that the specific decrease of the diffuser pressure corresponding to the continuous forcing was predicted correctly.     

Turbulent shear stress profiles in a bubbly channel flow assessed by particle tracking velocimetry

Particle tracking velocimetry (PTV) is applied to a bubbly two-phase turbulent flow in a horizontal channel at Re = 2 × 10⁴ to investigate the turbulent shear stress profile which had been altered by the presence of bubbles. Streamwise and vertical velocity components of liquid phase are obtained using a shallow focus imaging method under backlight photography. The size of bubbles injected through a porous plate in the channel ranged from 0.3 to 1.5 mm diameter, and the bubbles show a significant backward slip velocity relative to liquid flow. After bubbles and tracer particles are identified by binarizing the image, velocity of each phase and void fraction are profiled in a downstream region. The turbulent shear stress, which consists of three components in the bubbly two-phase flow, is computed by analysis of PTV data. The result shows that the fluctuation correlation between local void fraction and vertical liquid velocity provides a negative shear stress component which promotes frictional drag reduction in the bubbly two-phase layer. The paper also deals with the source of the negative shear stress considering bubble’s relative motion to liquid.     

Instability of the separated shear layer in flow past a cylinder: Forced excitation

The receptivity of the separated shear layer for Re = 300 flow past a cylinder is investigated by forced excitation via an unsteady inflow. In order to isolate the shear layer instability, a numerical experiment is set up that suppresses the primary wake instability. Computations are carried out for one half of the cylinder, in two dimensions. The flow past half a cylinder with steady inflow is found to be stable for Re = 300. However, an inlet flow with pulsatile perturbations, of amplitude 1% of the mean, results in the excitation of the shear layer mode. The frequency of the perturbation of the inlet flow determines the frequency associated with the shear layer vortices. For a certain range of forced frequencies the recirculation region undergoes a low‐frequency longitudinal contraction and expansion. An attempt is made to relate this instability to a global mode of the wake determined from a linear stability analysis. Interestingly, this phenomenon disappears when the outflow boundary of the computational domain is shifted sufficiently downstream. This study demonstrates the need of carefully investigating the effect of the location of outflow boundaries if the computational results indicate the presence of low‐frequency fluctuations. The effect of Re and amplitude of unsteadiness at the inlet are also presented. All computations have been carried out using a stabilized finite element formulation of the incompressible flow equations. Copyright © 2007 John Wiley & Sons, Ltd.     

Nanopositioning control of an electrostatic MEMS actuator: adaptive terminal sliding mode control approach

Nonlinear Dynamics - This research proposes an adaptive terminal sliding mode control strategy dedicated to motion tracking control of an electrostatic-actuated nanopositioning system. The...     

Turbulent pipe flow manipulation: some experimental and computational results for single manipulator rings

The effect on developing turbulent pipe flow of blade manipulator devices has been investigated both experimentally and computationally with a view towards reducing pipeline transmission losses. Wall pressure and mean axial velocity profiles have been obtained for nine manipulator configurations. Numerical calculations have been performed for two of these manipulator configurations using a finite volume method that incorporates a low Reynolds number model of turbulence. The results suggest that nett drag reduction may be possible in this flow although at present this seems unlikely. It remains to be answered whether any benefit can be achieved in fully developed flow for which both experimental and computational studies are now underway.     

遗传算法优化的神经网络在SINS/GPS中的应用

在不增加辅助系统的情况下,针对捷联惯性导航/全球定位组合导航系统(SINS/GPS)在GPS信号不可用时,其定位精度产生较大退化的问题,提出了遗传算法优化的径向基函数神经网络辅助组合导航系统定位的方法。当GPS信号可用时,采用遗传算法对径向基函数神经网络进行优化训练;当GPS信号不可用时,利用遗传算法优化后的径向基函数神经网络预测卡尔曼滤波器量测输入中的速度误差信息,使得卡尔曼滤波器能够继续工作并提供速度校正量。跑车实验表明,通过对速度进行误差补偿能够有效地修正位置误差,以GPS信号断开180 s的结果作分析,纯SINS模式的东向和北向位置误差分别为35.1 m和38.8 m,而本文所提方法的误差分别为10.5 m和7.2 m,其定位精度提高较为显著。     

Quantification of the effect of surface heating on shock wave modification by a plasma actuator in a low-density supersonic flow over a flat plate

This paper describes experimental and numerical investigations focused on the shock wave modification induced by a dc glow discharge. The model is a flat plate in a Mach 2 air flow, equipped with a plasma actuator composed of two electrodes. A weakly ionized plasma was created above the plate by generating a glow discharge with a negative dc potential applied to the upstream electrode. The natural flow exhibited a shock wave with a hyperbolic shape. Pitot measurements and ICCD images of the modified flow revealed that when the discharge was ignited, the shock wave angle increased with the discharge current. The spatial distribution of the surface temperature was measured with an IR camera. The surface temperature increased with the current and decreased along the model. The temperature distribution was reproduced experimentally by placing a heating element instead of the active electrode, and numerically by modifying the boundary condition at the model surface. For the same surface temperature, experimental investigations showed that the shock wave angle was lower with the heating element than for the case with the discharge switched on. The results show that surface heating is responsible for roughly 50 % of the shock wave angle increase, meaning that purely plasma effects must also be considered to fully explain the flow modifications observed.     

Duty cycle and directional jet effects of a plasma actuator on the flow control around a NACA0015 airfoil

This paper reports on the effects of a series of fluid-dynamic dielectric barrier discharge plasma actuators on a NACA0015 airfoil at high angle of attack. A set of jet actuators able to produce plasma jets with different directions (vectoring effect) and operated at different on/off duty cycle frequencies are used. The experiments are performed in a wind tunnel facility. The vectorized jet and the transient of the flow induced by unsteady duty cycle operation of each actuator are examined and the effectiveness of the actuator to recover stall condition in the range of Reynolds numbers between 1.0 × 10⁵ and 5.0 × 10⁵ (based on airfoil chord), is investigated. The actuator placed on the leading edge of the airfoil presents the most effective stall recovery. No significant effects can be observed for different orientations of the jet. An increase of the stall recovery is detected when the actuator is operated in unsteady operation mode. Moreover, the frequency of the on/off duty cycle that maximizes the stall recovery is found to be a function of the free stream velocity. This frequency seems to scale with the boundary layer thickness at the position of the actuator. A lift coefficient increase at low free stream velocities appears to linearly depend on the supply voltage.