On-Line Black-Box Aerodynamic Performance Optimization for a Morphing Wing With Distributed Sensing and Control

Abstract

Inspired by nature, smart morphing technologies enable the aircraft of tomorrow to sense their environment and adapt the shape of their wings in flight to minimize fuel consumption and emissions. A primary challenge on the road to this feature is how to use the knowledge gathered from sensory data to establish an optimal shape adaptively and continuously in flight. To address this challenge, this article proposes an online black-box aerodynamic performance optimization architecture for active morphing wings. The proposed method integrates a global online-learned radial basis function neural network (RBFNN) model with an evolutionary optimization strategy, which can find global optima without requiring in-flight local model excitation maneuvers. The actual wing shape is sensed via a computer vision system, while the optimized wing shape is realized via nonlinear adaptive control. The effectiveness of the optimization architecture was experimentally validated on an active trailing-edge (TE) camber morphing wing demonstrator with distributed sensing and control in an open jet wind tunnel. Compared with the unmorphed shape, a 7.8% drag reduction was realized, while achieving the required amount of lift. Further data-driven predictions have indicated that up to 19.8% of drag reduction is achievable and have provided insight into the trends in optimal wing shapes for a wide range of lift targets.