Cluster-based feedback control of turbulent post-stall separated flows

Abstract

We propose a novel model-free self-learning cluster-based control strategy for general nonlinear feedback flow control technique, benchmarked for high-fidelity simulations of post-stall separated flows over an airfoil. The present approach partitions the flow trajectories (force measurements) into clusters, which correspond to characteristic coarse-grained phases in a low-dimensional feature space. A feedback control law is then sought for each cluster state through iterative evaluation and downhill simplex search to minimize power consumption in flight. Unsupervised clustering of the flow trajectories for in-situ learning and optimization of coarse-grained control laws are implemented in an automated manner as key enablers. Re-routing the flow trajectories, the optimized control laws shift the cluster populations to the aerodynamically favorable states. Utilizing limited number of sensor measurements for both clustering and optimization, these feedback laws were determined in only $O(10)$ iterations. The objective of the present work is not necessarily to suppress flow separation but to minimize the desired cost function to achieve enhanced aerodynamic performance. The present control approach is applied to the control of two and three-dimensional separated flows over a NACA 0012 airfoil with large-eddy simulations at an angle of attack of $9^{\circ }$, Reynolds number $Re=23,000$ and free-stream Mach number $M_{\infty }=0.3$. The optimized control laws effectively minimize the flight power consumption enabling the flows to reach a low-drag state. The present work aims to address the challenges associated with adaptive feedback control design for turbulent separated flows at moderate Reynolds number.

arXiv comment32 pages, 18 figures

References in corpus (10)

• Discovering governing equations from data: Sparse identification of nonlinear dynamical systems
• Community structure in directed networks
• Artificial Neural Networks trained through Deep Reinforcement Learning discover control strategies for active flow control
• Cluster-based reduced-order modelling of a mixing layer
• Sparse reduced-order modeling : Sensor-based dynamics to full-state estimation
• Closed-loop separation control using machine learning
• Resolvent-analysis-based design of airfoil separation control
• Closed-loop separation control over a sharp edge ramp using Genetic Programming
• Phase-response analysis of synchronization for periodic flows
• Networked oscillator based modeling and control of unsteady wakes

Cited by in corpus (19)

• Artificial intelligence control of a turbulent jet
• Data-driven modelling of nonlinear spatio-temporal fluid flows using a deep convolutional generative adversarial network
• Deep Model Predictive Control with Online Learning for Complex Physical Systems
• A review on turbulent and vortical flow analyses via complex networks
• Cluster-based network model
• Machine learning flow control with few sensor feedback and measurement noise
• Stabilization of the fluidic pinball with gradient-enriched machine learning control
• Cluster-based hierarchical network model of the fluidic pinball -- Cartographing transient and post-transient, multi-frequency, multi-attractor behaviour
• Phase-based control of periodic fluid flows
• Stabilization of a multi-frequency open cavity flow with gradient-enriched machine learning control
• Trajectory-optimized cluster-based network model for the sphere wake
• Randomized methods to characterize large-scale vortical flow network
• Load estimation in unsteady flows from sparse pressure measurements: Application of transition networks to experimental data
• Cycle-to-cycle variations in cross-flow turbine performance and flow fields
• Self-tuning model predictive control for wake flows
• Augmenting a Physics-Informed Neural Network for the 2D Burgers Equation by Addition of Solution Data Points
• xMLC -- A Toolkit for Machine Learning Control
• Network-based analysis of fluid flows: Progress and outlook
• Towards human-interpretable, automated learning of feedback control for the mixing layer