Bulletin of the American Physical Society
72nd Annual Meeting of the APS Division of Fluid Dynamics
Volume 64, Number 13
Saturday–Tuesday, November 23–26, 2019; Seattle, Washington
Session Q27: Flow Control: Model Reduction |
Hide Abstracts |
Chair: Aditya Nair, University of Washington Room: 609 |
Tuesday, November 26, 2019 7:45AM - 7:58AM |
Q27.00001: Energy-optimal phase control of unsteady fluid flows Aditya Nair, Kunihiko Taira, Bingni Brunton, Steven Brunton Controlling the phase of oscillation of unsteady fluid flows is crucial in terms of advancing and delaying flow field characteristics. Our goal is to design an optimal flow control strategy that alters the oscillation phase in fluid flows using minimum energy input. We perform a phase reduction analysis to construct a reduced-order model in terms of a phase-sensitivity function that tracks the phase of oscillation in response to impulse perturbations. Using the phase sensitivity function and its gradient, the optimal control law is obtained by solving the Euler-Lagrange equations as a two-point boundary value problem. We demonstrate the approach for incompressible flow over a cylinder at low Reynolds number and discuss extensions to complex flows. Multiple actuation strategies based on blowing and rotary control are also explored. Finally, we examine the synchronization and desynchronization properties of the optimally controlled flow and discuss broader implications for flow control. [Preview Abstract] |
Tuesday, November 26, 2019 7:58AM - 8:11AM |
Q27.00002: Spectral POD analysis of low Reynolds flow past finite cylinders Matteo Chiatto, Caroline Cardinale, Jessica K. Shang, Luigi de Luca, Francesco Grasso The need to understand the physics and to control a flow field has led scientists to develop techniques suitable for characterizing the dynamics and the topology of the flow based on simplified equations, and by exploiting experimental and/or numerical data. Major analysis techniques include POD, DMD, Spectral POD (SPOD; Towne et al. JFM 2018, 847). SPOD has the ability to represent structures evolving coherently in space and time, and it is here applied to study the flow past finite oblique cylinders. The study is aimed at investigating shedding regimes at various low Reynolds numbers. The wake behind the cylinder is experimentally investigated by means of a digital visualization technique described by Shang et al. (JFM 2018, 837). The major results consist in extracting the wake dynamics and identifying the eigen-directions associated with the most energetic structures. The wake field is reconstructed by considering just the few selected modes. Such a low dimensional decomposition is able to capture the dominant shedding modes, by comparing both image reconstruction and quantitative shedding Strouhal numbers to previous temporal Fourier transform analysis, at various Reynolds numbers. [Preview Abstract] |
Tuesday, November 26, 2019 8:11AM - 8:24AM |
Q27.00003: Model-Based Estimation of Vortex Shedding in Unsteady Cylinder Wakes Jiwen Gong, Jason Monty, Simon Illingworth This work considers single-sensor based estimation of vortex shedding in cylinder wakes at $Re = 100$ in simulations and at $Re = 1036$ in experiments. A model based on harmonic decomposition is developed to capture the periodic dynamics of vortex shedding. Two model-based methods are proposed to estimate time-resolved flow fields. First, Linear Estimation (LE) which implements a Kalman Filter to estimate the flow. Second, Linear-Trigonometric Estimation (LTE), which utilizes the same Kalman Filter together with a nonlinear relationship between harmonics of the vortex shedding frequency. LTE shows good performance and outperforms LE in reconstructing vortex shedding. Physically this suggests that, at the Reynolds numbers considered, the higher harmonic motions are slave to the fundamental frequency. [Preview Abstract] |
Tuesday, November 26, 2019 8:24AM - 8:37AM |
Q27.00004: The dynamical states of a circular cylinder wake influenced by a leeward control rod Murilo Cicolin, Peter Bearman, Oliver Buxton, Gustavo Assi Experiments were carried out in a recirculating water channel with a circular cylinder fitted with one control rod at different positions. The rod had a diameter 10 times smaller than the main cylinder. The distance between the centre of cylinders was kept constant (R/D=0,7) and the angle between them varied from 40 to 90 degrees, when at 0 degrees the control cylinder is placed in the wake on a line passing through the front stagnation point of the main cylinder and its centre. Reynolds number was 20,000 for all cases, based on the main cylinder’s diameter. High-resolution PIV fields were acquired focusing on the two-dimensional wake of both cylinders. Results showed that for all cases the control rod induced a significant change in the wake when compared to the bare cylinder. Three main dynamical states were identified according to the position of control rod across the shear layer: immersed in the recirculation region, in the middle of shear layer or outside of it. Moreover, a bistable transitional state was identified between the first and second states. The second state shows a suppression of vortex shedding and a significant reduction of drag forces for the combined system. [Preview Abstract] |
Tuesday, November 26, 2019 8:37AM - 8:50AM |
Q27.00005: Koopman mode representations of vortex dynamics in viscous flows Ke-Chu Lee, Sam Kaufman-Martin, Samaneh Sadri, Poorva Shukla, Igor Mezić, Paolo Luzzatto-Fegiz Vortex dynamics plays an important role in transitional and turbulent flows, where instabilities introduce important criteria for safety and performance of systems like turbomachinery and aerospace vehicles. In order to control these vortical flows, accurate, general and efficient models of vortex dynamics are needed. Here we explore the ability of Koopman mode decomposition (KMD) to provide such models and uncover physical mechanisms. Using pseudo-spectral simulations, we consider vortex dynamics in viscous flow for different Reynolds numbers and different initial vortex geometries. We start with co-rotating vortex pairs, for which we find that KMD modes and eigenvalues are in close agreement with linear stability predictions. We find that the onset of symmetry-breaking is weakly sensitive to Reynolds number, but highly sensitive to small details in the initial conditions. We provide a detailed comparison of the relative abilities of KMD and proper orthogonal decomposition (POD) to capture this behavior. Finally, we consider implications for modeling and controlling more general vortex flows. [Preview Abstract] |
Tuesday, November 26, 2019 8:50AM - 9:03AM |
Q27.00006: Dynamic mode analysis and control of vortical flows in pump sumps Byungjin An, Qiong Liu, Motohiko Nohmi, Masashi Obuchi, Kunihiko Taira Pump sumps are settling chambers for incoming flows prior to their removal by pumps. These pump sumps are widely used for drainage in pumping stations and power plants. During off-design operations, free surface and sub-surface vortices often appear in pump sumps, causing significant pump performance degradation and system vibrations. Modal analysis is performed to obtain insights to develop an effective and highly robust suppressing device of vortex for a scaled pump sump. Dynamic mode decomposition (DMD) is used to extract the dynamic features with respect to the vortical flow obtained from the large eddy simulation (LES). The dominant DMD modes are useful for understanding the complex vortex dynamics and developing a novel control device for the suppression of vortex formation. In addition to examining the complex flow in the pump sump, we also consider a wall-normal vortex model. Active flow control is considered for the attenuation of such vortex. The resulting increase in pressure at the vortex core suppresses the detrimental effects. A comparative analysis of steady and unsteady actuation methods is carried out through LES. The characteristics of the successful control mechanisms will be identified and discussed. [Preview Abstract] |
Tuesday, November 26, 2019 9:03AM - 9:16AM |
Q27.00007: A Hybrid Low-order Model of Dynamic Lift Response to Time-varying Actuation Xuanhong An, David Willams Time-varying actuation from a synthetic jet actuator has previously been shown to be an effective way of controlling unsteady flow separation. In order to integrate this type of actuation into a real-time control system, a low-order model of the lift response to the actuation is desired. The current work proposes a low-order approach to modeling the lift response to time-varying leading edge actuation on a stalled airfoil. Dynamic Mode Decomposition (DMD) is employed to extract from the flowfield the critical dynamical information, which is connected to the lift variation. It is shown that there are two sets of DMD modes associated with two different frequencies. These two frequencies are contained in both the actuation signal and the lift response. Based on the information provided by the critical DMD modes, we propose a hybrid low-order model which consists of a time delay decay model and a convolution integral model. This hybrid model is capable of accurately predicting the lift response of an airfoil to the time-varying actuation. [Preview Abstract] |
Tuesday, November 26, 2019 9:16AM - 9:29AM |
Q27.00008: Modeling the transient response to momentum injection for flow over an airfoil in deep stall using data-driven and projection-based methods Katherine Asztalos, Scott Dawson, David Williams Direct numerical simulations are performed for leading-edge momentum injection control of flow over a NACA0009 airfoil at post-stall angles of attack. We consider cases where the wake is both stable (Re = 200) and unstable (Re = 500), and find that the flow response is comparable in both regimes, though with some sensitivity to the instantaneous flow state in the unstable case. We investigate the physical mechanisms governing the interactions between the disturbance and the natural flow state using both dynamic mode decomposition (which is an entirely data-driven approach), and Galerkin projection of the governing Navier-Stokes equations onto a subspace obtained from proper orthogonal decomposition. In particular, we show that projecting a linearized Galerkin projection model can accurately predict DMD modes, giving an understanding of the origin of these modes from the underlying governing equations. More generally, this also suggests a method for approximating DMD modes from non-time-resolved data. We assess the ability of the resulting models to predict the time-evolution of the flow state of the actuated system. We also study the ability of our models to capture the finite-time-horizon energy growth present in the full system. [Preview Abstract] |
Tuesday, November 26, 2019 9:29AM - 9:42AM |
Q27.00009: Effect of spanwise wall oscillations on dynamics and evolution of near-wall coherent structures in a turbulent pipe flow at low and moderate Reynolds numbers Daniel Coxe, Ronald Adrian, Yulia Peet Presented is the temporal evolution of a single conditional hairpin vortex in a turbulent pipe flow with transversely oscillated walls at low and moderate Reynolds numbers, compared to a baseline non-oscillated turbulent pipe flow case. A conditional hairpin is generated by a Linear Stochastic Estimation of a fluctuating velocity field of a non-oscillated pipe flow, obtained via Direct Numerical Simulations with a spectral-element method. An extracted hairpin is placed as an initial condition into the flow of interest and is convected downstream by a turbulent mean velocity profile. A goal of the study is to investigate the effect of wall oscillations on the development and growth of a conditional hairpin and the mechanisms associated with a potential suppression of auto-generation in a drag reduced flow. Subsequent formation and evolution of secondary and tertiary hairpins and how these processes are modified by a spanwise wall oscillation are also documented. A spatially averaged Reynolds stress profile along with the wall shear stress are presented to quantify the effects of auto-generation. [Preview Abstract] |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700