Aerodynamic Parameter Prediction via Artificial Hair Sensors with Signal Power in Turbulent Flow

摘要

Spatiotemporal surface flow information obtained from distributed arrays of bioinspired hair sensors are capable of predicting the real-time aerodynamic parameters (i.e., lift, moment, angle of attack, and freestream velocity) on a representative wing section. When combined with an appropriate model of the system, these sensors can provide vital information about gust disturbance as well as state estimation of an aeroelastic system, both of which are essential for effective vibration suppression control system design. In a typical aeroelastic system undergoing a periodic change in bending and torsion motion, a number of hair sensors can be in turbulent flow regions at any instant, resulting in random vibration response. This paper specifically investigates the effect in aerodynamic parameter prediction when sensor measurements from both laminar and turbulent flow regions are combined. It also investigates the idea of incorporating the sensor signal power along with sensor information to increase the prediction accuracy in such a scenario. The experimental results show that incorporating sensor response from both laminar and turbulent flow regions improves the prediction for the whole operating region containing both positive and negative angle of attack. Moreover, incorporating the signal power further improves the prediction accuracy and precision.