Experimental and Numerical Study of Plasma Based Flow Control for Low Pressure Gas Turbines Operating at Low Reynolds Numbers

High altitude cruise represents a crucial issue for small size low pressure turbines (LPT), commonly used in the propulsion of unmanned air vehicles (UAVs). The Reynolds number can drop below 25000, which in turn can lead to laminar boundary layer separation on the suction surface of the blades. This makes the turbine working in off-design conditions with very poor performances. Modifying the blade shape to counteract the boundary layer separation is not a feasible solution since the performance of the turbine will be adversely affected at the engine design conditions (take-off and landing). Therefore, the implementation of a boundary layer control system on the suction side of the turbine able to operate only at low Reynolds number is the most practical solution. The present study investigates experimentally and numerically the potential of an alternate current (AC) driven Single Dielectric Barrier Discharge Plasma Actuator (AC-SDBDPA) to reattach the separated flow at a Reynolds number around 2·104. The SDBDPA was designed and manufactured by means of lithographic technique, which ensured a thin metal deposition with high manufacturing reliability control. The experimental approach comprised the actuator testing over a curved plate with a shape designed to reproduce the suction surface of a LPT. A closed loop wind tunnel was employed. The curved plate was mounted directly over the bottom wall of the test section. The AC-SDBDPA was placed in a grove made at the middle of the curved plate and located at the front side of the adverse pressure gradient region. Sinusoidal voltage excitation was tested. The flow measurements –with and without actuation– were carried out by laser Doppler velocimetry (LDV) and particle image velocimetry (PIV). Planar measurements were performed over the curved plate at the midspan plane. Simultaneously to the velocity measurements the applied voltage and the discharge current were acquired in order to determine the device dissipated power. The experimental data was complemented with CFD simulations based on the finite volume method. The actuator effect was modelled as a time-constant body force calculated prior to the fluid flow simulation by using a dual potential algebraic model. Reynolds Averaged Navier Stokes (RANS) method was used to consider the turbulence effect. The validity of the numerical model allows to expand the study of the actuation effect including different locations and multiple devices, saving considerably experimental efforts