Aerodynamic Parameter Prediction on a Airfoil with Flap via Artificial Hair Sensors and Feedforward Neural Network

Abstract

Gust alleviation and flutter suppression are essential elements of an effective fly-by-feel system. Knowledge of real-time forces and moments can have huge effect on designing an effective controller for flutter suppression and gust rejection. One unique method of predicting forces and moments is to use distributed arrays of artificial hair sensors that are capable of sensing the environment and therefore capturing important flow features. In this paper, the local flow measurement from the artificial hair sensor is used with feed-forward neural network to predict the aerodynamic parameters (angle of attack, freestream velocity, lifte coefficient and moment coefficient per unit span, and flap angle) on an airfoil containing control surface. These aerodynamic parameters can be combined with the airfoil’s physical parameters to predict the real time lift and moment. Also, the effect of artificial hair sensor integration location on prediction of aerodynamic parameters is studied.