arXiv subjects

Nirmal J. Nair

Publications and source records attributed to Nirmal J. Nair.

4 recordsLinked to original sources

arXiv2022

Bio-inspired variable-stiffness flaps for hybrid flow control, tuned via reinforcement learning

A bio-inspired, passively deployable flap attached to an airfoil by a torsional spring of fixed stiffness can provide significant lift improvements at post-stall angles of attack. In this work, we describe a hybrid active-passive variant to this purely passive flow control paradigm, where the stiffness of the hinge is actively varied in time to yield passive fluid-structure interaction (FSI) of greater aerodynamic benefit than the fixed-stiffness case. This hybrid active-passive flow control strategy could potentially be implemented using variable stiffness actuators with less expense compared with actively prescribing the flap motion. The hinge stiffness is varied via a reinforcement learning (RL)-trained closed-loop feedback controller. A physics-based penalty and a long-short-term training strategy for enabling fast training of the hybrid controller are introduced. The hybrid controller is shown to provide lift improvements as high as 136\% and 85\% with respect to the flap-less airfoil and the best fixed-stiffness case, respectively. These lift improvements are achieved due to large-amplitude flap oscillations as the stiffness varies over four orders of magnitude, whose interplay with the flow is analyzed in detail.

physics.flu-dyn↗

arXiv2022

Fluid-structure interaction of a bio-inspired passively deployable flap for lift enhancement

Birds have a remarkable ability to perform complex maneuvers at post-stall angles of attack. The passive deployment of self-actuating covert feathers in response to unsteady flow separation while performing such maneuvers provides a passive flow control paradigm for these aerodynamic capabilities. Most studies involving covert-feathers-inspired passive flow control have modeled the feathers as a rigidly attached or a freely moving flap on a wing. A flap mounted via a torsional spring enables a configuration more emblematic of the finite stiffness associated with the covert-feather dynamics. The performance benefits and flow physics associated with this more general case remain largely unexplored. In this work, we model covert feathers as a passively deployable, torsionally hinged flap on the suction surface of a stationary airfoil. We numerically investigate this airfoil-flap system at a low Reynolds number of $Re=1{,}000$ and angle of attack of $20^\circ$ by performing high-fidelity nonlinear simulations using a projection-based immersed boundary method. A parametric study performed by varying the stiffness of the spring, mass of the flap and location of the hinge yielded lift improvements as high as 27% relative to the baseline flap-less case and revealed two dominant flow behavioral regimes. A detailed analysis revealed that the stiffness-dependent mean flap deflection and inertia-dependent amplitude and phase of flap oscillations altered the dominant flow characteristics in both the regimes. Of special interest for performance benefits were the flap parameters that enhanced the lift-conducive leading-edge vortex while weakening the trailing-edge vortex and associated detrimental effect of upstream propagation of reverse flow. These parameters also yielded a favorable temporal synchronization of flap oscillations with the vortex-shedding process in both regimes.

physics.flu-dyn↗

arXiv2019

Leveraging reduced-order models for state estimation using deep learning

State estimation is key to both analyzing physical mechanisms and enabling real-time control of fluid flows. A common estimation approach is to relate sensor measurements to a reduced state governed by a reduced-order model (ROM). (When desired, the full state can be recovered via the ROM). Current methods in this category nearly always use a linear model to relate the sensor data to the reduced state, which often leads to restrictions on sensor locations and has inherent limitations in representing the generally nonlinear relationship between the measurements and reduced state. We propose an alternative methodology where a neural network architecture is used to learn this nonlinear relationship. Neural network is a natural choice for this estimation problem, as a physical interpretation of the reduced state-sensor measurement relationship is rarely obvious. The proposed estimation framework is agnostic to the ROM employed, and can be incorporated into any choice of ROMs derived on a linear subspace (e.g., proper orthogonal decomposition) or a nonlinear manifold. The proposed approach is demonstrated on a two-dimensional model problem of separated flow around a flat plate, and is found to outperform common linear estimation alternatives.

physics.flu-dyn↗

arXiv2019

Transported snapshot model order reduction approach for parametric, steady-state fluid flows containing parameter dependent shocks

A new model order reduction approach is proposed for parametric steady-state nonlinear fluid flows characterized by shocks and discontinuities whose spatial locations and orientations are strongly parameter dependent. In this method, solutions in the predictive regime are approximated using a linear superposition of parameter dependent basis. The sought after parametric reduced-basis are obtained by transporting the snapshots in a spatially and parametrically dependent transport field. Key to the proposed approach is the observation that the transport fields are typically smooth and continuous, despite the solution themselves not being so. As a result, the transport fields can be accurately expressed using a low-order polynomial expansion. Similar to traditional projection-based model order reduction approaches, the proposed method is formulated mathematically as a residual minimization problem for the generalized coordinates. The proposed approach is also integrated with well-known hyper-reduction strategies to obtain significant computational speed-ups. The method is successfully applied to the reduction of a parametric 1-D flow in a converging-diverging nozzle, a parametric 2-D supersonic flow over a forward facing step and a parametric 2-D jet diffusion flame in a combustor.

physics.flu-dyn↗