Bulletin of the American Physical Society
77th Annual Meeting of the Division of Fluid Dynamics
Sunday–Tuesday, November 24–26, 2024; Salt Lake City, Utah
Session X33: Flow Control IV: General |
Hide Abstracts |
Chair: Thomas Bewley, UC San Diego Room: 255 E |
Tuesday, November 26, 2024 8:00AM - 8:13AM |
X33.00001: Abstract Withdrawn
|
Tuesday, November 26, 2024 8:13AM - 8:26AM |
X33.00002: A regularised shadowing method for sensitivity analysis of turbulence flows Liang Fang, George Papadakis There is increasing demand for optimisation using scale-resolving methods, such as direct and large-eddy simulations. However, standard methods to evaluate the gradient of a time-averaged objective function to a control parameter, for example the adjoint method, fail for chaotic turbulent flows due to the so-called “butterfly effect”. A promising method for calculating the gradient (or sensitivity) is the Least-Squares Shadowing (LSS) method. However, for systems that are not uniformly hyperbolic, the resulting system matrix has a very large condition number, which affects the accuracy of the results. To address this challenge, we introduce a new regularised non-intrusive LSS method. By applying Tikhonov regularisation, we can make the system well-conditioned. The optimal regularisation parameter is obtained by finding the maximum curvature of the L-curve. This innovation improves the accuracy of the method and significantly reduces the standard deviation of the computed gradient. It therefore extends the applicability of LSS to a broader range of turbulent flow systems. |
Tuesday, November 26, 2024 8:26AM - 8:39AM |
X33.00003: Model-based reinforcement learning for chaotic flow control Priyam Gupta, Max Weissenbacher, Georgios Rigas Deep reinforcement learning (DRL) has demonstrated significant potential for improving the efficiency of fluid-based systems. However, numerical solvers for high-dimensional, complex flow environments require substantial computational resources, often intractable. This is especially problematic for DRL, which typically requires numerous interactions with the environment to achieve convergence. To address sample efficiency, many model-based reinforcement learning (MBRL) methods have been proposed. However, these algorithms are primarily developed and benchmarked on classical control problems and game-like environments, which are non-chaotic. Existing research on applying MBRL to flow problems has focused mainly on the weakly chaotic regime. Consequently, the effectiveness of MBRL for highly chaotic systems remains unknown, as most real-world flow control problems involve chaotic turbulent flows. In this study, we conduct a comprehensive investigation to determine the chaotic regimes where MBRL is beneficial and identify the key parameters required to achieve optimal controller performance for fluid flows. We highlight pitfalls in using MBRL for controlling chaotic systems and propose suitable adaptations to enhance their performance. Additionally, we compare MBRL with state-of-the-art model-free algorithms. |
Tuesday, November 26, 2024 8:39AM - 8:52AM |
X33.00004: Flow control of a turbulent separation bubble: Information-theoretic approach Tristan Villanueva, Gonzalo Arranz, Adrian Lozano-Duran Turbulent separation bubbles (TSBs) may form over aerodynamic surfaces under the influence of adverse pressure gradients, significantly impacting aerodynamic efficiency. Thus, minimizing these bubbles is crucial for enhancing the performance and reliability of aircraft. We investigate the effectiveness of a zero-net-mass-flux (ZNMF) control strategy for a TSB within an information-theoretic framework (IT-control). The control problem is framed in information-theoretic terms by envisioning the tandem sensor-actuator as a device that reduces the unknown information about the state to be controlled. By leveraging the information-theoretic formulation, we determine the optimal (1) locations of sensors, (2) locations of actuators, and (3) actuator parameters that minimize the error between the actual and desired probability distributions of the recirculation length of the turbulent bubble. The IT-control approach is compared to other ZNMF controllers. |
Tuesday, November 26, 2024 8:52AM - 9:05AM |
X33.00005: Lift generation on a sphere using active surface morphing Putu Brahmanda Sudarsana, Rodrigo Vilumbrales-Garcia, Anchal Sareen This study investigates the generation of lift on a sphere by pneumatically manipulating its surface topology with an asymmetric roughness distribution using morphable dimples. The experiments were conducted at Reynolds numbers of 6×104 ≤ Re ≤ 1.3×105 and dimple depth ratios of 0 ≤ k/d ≤ 1×10-2. A generation of lift forces up to 80% of the drag was found, comparable to those produced by the Magnus Effect. The dimple depth ratio affects both the Re at which lift generation begins and the maximum lift coefficient CL achievable. The optimal k/d for maximum lift varies with Re where low Re requires deeper dimples and shallower dimples for high Re. For a fixed Re, increasing k/d monotonically increases lift until a critical k/d is reached, beyond which lift decreases. Particle Image Velocimetry (PIV) revealed a delay in flow separation on the dimpled side while the smooth side remains unchanged, resulting in asymmetric boundary layer separation that leads to lift generation. Beyond the critical k/d, the flow separation location moves upstream as the rear wake increases. This study establishes the foundation for flow separation control over bluff bodies and paves the way for maneuvering applications. |
Tuesday, November 26, 2024 9:05AM - 9:18AM |
X33.00006: High-Fidelity Simulations of Single and Tandem Perforated Elastic Vortex Generators Karan Kakroo, Hamid Sadat Vortex generators (VGs) are essential for enhancing mixing across various industrial sectors. Research on Rigid Vortex Generators (RVGs) has shown significant improvements in mixing but with notable pressure drops. Flexible or elastic vortex generators (EVGs) offer enhanced performance with lower pressure drops, yet further innovations are needed to reduce pressure drops further while maintaining efficiency. This study explores the use of single and tandem perforated EVGs (PEVGs) to achieve a more efficient vortex generator. High-fidelity 2-way coupled Fluid Structure Interaction (FSI) simulations are conducted to investigate the response of PEVGs across a wide range of dimensionless rigidity, mass ratios, Reynolds numbers, and porosity levels. Additional simulations for non-perforated EVGs are conducted for comparison and validation against available experimental data. The dynamics of PEVGs are examined by analyzing response characteristics such as inclination angle, phase portrait, and response harmonics. Local flow dynamics are also studied by investigating vortical structures and velocity fields. Based on the defined parameters, various flexibility modes and patterns are observed, and the wakes associated with these different modes are compared and analyzed for both perforated and non-perforated EVGs, as well as for the single and tandem configurations. |
Tuesday, November 26, 2024 9:18AM - 9:31AM |
X33.00007: Computational Simulation of Wave Generation in Flexible Structures using Piezoelectric Actuators for Flow Control Uchenna Emmanuel Ogunka, Iman Borazjani
|
Tuesday, November 26, 2024 9:31AM - 9:44AM |
X33.00008: A 3D Comparison of Sprayed Liquid Flap Multiphase Modeling at Activation Alexander Spitzer, George Loubimov, Michael Kinzel Powered-lift concepts are often used to increase aerodynamic lift in specific scenarios like aircraft takeoff and landing or reduce lift during wind turbine high load conditions. A Sprayed Liquid Flap (SLF) is a novel powered-lift concept previously shown (computationally) to improve aerodynamic performance. Similar to jet flaps, fluid is expelled from the pressure surface, providing both increased lift and increased aerodynamic performance. Modeling SLFs requires coupled multiphase physics as each phase directly impacts not only each other, but also the aerodynamic forces on the airfoil. Previous 2D-SLF research efforts suggest aerodynamic performance can be well estimated using mid-fidelity Euler-Euler approaches. However, unsteady 3D aerodynamics of SLF activation have yet to be explored. This research effort focuses on comparing 3D, unsteady Euler-Euler multiphase methods against a similar Euler-Lagrange counterpart. Specifically, the effort seeks to understand how Euler-Euler models with particle drag-force coupling compares to Euler-Lagrange models with additional physics by including drag force coupling as well as particle collisions, coalescence, deformation, and breakup models. Results will be summarized and evaluated for aerodynamics as well as details of the multiphase flow will be explored. |
Tuesday, November 26, 2024 9:44AM - 9:57AM |
X33.00009: Blowin' in the wind Thomas R Bewley Most at APS-DFD are well familiar with the Tacoma Narrows Bridge disaster. The operation of tethered observation balloons often suffer from similar vortex-induced oscillations, even at modest wind velocities. This talk will discuss two new feedback control techniques to subdue such wind-induced oscillations. In the case of operation with a single tether, a simple nonlinear (period doubling) feedback control strategy, synchronizing the motion of the winch (on the ground) controlling the tether with the wind-induced oscillations of the balloon, in a manner roughly opposite to what a child learns to do in order to drive up the oscillations of a playground swing, is found to be effective. In the case of operation with multiple (>6) tethers, analysis in a tensegrity-based framework is found to be effective, and a simple linear program (LP) may be formulated and solved to minimize the maximum tension, and to maximize the minimum tension, of the several tethers, thereby at any moment providing the most suitable set of tensions (driven by a set of winches) to stabilize the balloon. Ongoing efforts by USAFA cadets to validate such solutions experimentally will also be discussed. |
Tuesday, November 26, 2024 9:57AM - 10:10AM |
X33.00010: Experimental Investigation on the Interaction Between a Fluidic Oscillator Array and a Cross-flow Haiyang Hu, Tulasi R Vechalapu, Kota Mitsumoto, Pritesh Tiwari Active flow control (AFC) methods have been developed and widely studied for fluid mechanics in past decades. Among all the AFCs, the fluidic oscillators are one of the notable techniques that have been proven to yield significant efficiency improvements by preventing flow separation in various applications. While numerous researchers have explored the fluid oscillator, their focus has primarily been on the performance of single actuators. The fundamental knowledge of the multi-actuator interaction with the cross-flow, which is crucial to the optimization and reduce-order modeling of fluidic oscillator arrays, is yet limited. In the present study, an experimental investigation is conducted to evaluate the flow field characteristics of the oscillating jet array into free stream cross-flow. The actuator array is 3D printed using five oscillators with two feedback channels (i.e., sweeping jet actuator). Driven by a high-pressure air source, the actuator array is installed spanwise on a flat plate in a closed-circuit low-speed wind tunnel at the University of Alabama in Huntsville. Both the hotwire anemometer and the Particle Image Velocimetry (PIV) system are utilized in the experiments to quantify the fundamental flow behavior of a single actuator and mixing between different actuators at various mass flow rates of the actuator and different free-stream velocities. While the 2D PIV is used during the wind tunnel experiments to capture the streamwise mixing between emitting oscillating jet and cross-flow field, a stereoscopic PIV system is used to identify the spanwise vorticity structure between multiple actuators to further understand the mixing process between multiple actuators and cross flow. Our preliminary findings indicate that the operation of the multi-actuator array in the quiescent air can reduce the oscillation angle of every single actuator through the interaction between adjacent actuators. The mixing of the emitting jet takes place within the boundary layer. The stereoscopic PIV can clearly capture the vortex structure of the multiple jet interaction. This study aims to gain a more fundamental comprehension of the flow behavior when oscillating jets emit into the cross-flow, such that it provides more information for the flow control design in the future. |
Tuesday, November 26, 2024 10:10AM - 10:23AM |
X33.00011: Passive Flow Control on a Circular Cylinder Using a Small Spoiler – A Window Into the Effect of Baseball Seams Jack Elliott, Alex Nielson, Barton L Smith Recent research has shown that baseball seams behave as “spoilers” by promoting boundary-layer separation at the seam when those seams lie near a region of adverse pressure gradient (Smith et al., How seams alter boundary layer separation points on baseballs. Exp Fluids 65, 27 (2024)). This presentation will show the early results of a wind tunnel study in which a cylinder of similar size to a baseball (about 70 mm) is placed in the tunnel with similar velocity to a baseball pitch (about 40 m/s). The spoiler is isosceles-triangular-shaped, protruding 1 mm from the cylinder surface and 4 mm in the streamwise direction. The main measurements are the surface pressure distribution around the cylinder. The present study shows that when the spoiler is in the region 80º < θ < 120º, boundary-layer separation occurs at the spoiler, in advance of the position near θ = 120º where separation normally occurs for these turbulent boundary layers. A similar range was reported for baseballs. Further, the present study shows that location of the separation point, determined by the spoiler location, alters the pressure distribution around the cylinder. Specifically, the pressure upstream of the spoiler increases compared to the non-spoiler side, with this behavior increasing according to decreasing spoiler angle. However, the base pressure, separation on the non-spoiler side, and non-spoiler side pressure remain constant across observed spoiler locations. Thus, spoiler location determines the magnitude of lift and drag compared to the cylinder with no spoiler. |
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. |
© 2025 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