Metamodeling-based parametric optimization of DBD plasma actuation to suppress flow separation over a wind turbine airfoil model

Abstract

While dielectric-barrier-discharge (DBD) based plasma actuation systems have been successfully demonstrated to suppress massive flow separation over wind turbine blades to reduce the transient aerodynamic loadings acting on the turbine blades, it is still a non-trivial task to establish a best combination of various operating parameters for a DBD plasma actuation system to achieve the optimized flow control effectiveness. In the present study, a regression Kriging based metamodeling technique is developed to optimize the operating parameters of a DBD plasma actuation system for suppressing deep stall over the surface of a wind turbine blade section/airfoil model. The data points were experimentally obtained by embedding a nanosecond-pulsed DBD (NS-DBD) plasma actuator at the leading edge of the airfoil model. The applied voltage and frequency for the NS-DBD plasma actuation were used as the design variables to demonstrate the optimization procedure. The highest possible lift coefficient of the turbine airfoil model at deep stalled angles of attack (i.e., α = 22° and 24°) were selected as the objective function for the optimization. It was found that, while the metamodeling-based procedure could accurately predict the objective function within the bounds of the design variables with an uncertainty ~ 2%, a global accuracy level of ~ 97% was achieved within the whole design space.