FFECTIVE control of the aerodynamic loads on airfoils maneuvering over a wide range of incidence angles and with varying pitch rates is vital for the development of new generation aerodynamic systems. A classic example is the control of dynamic stall on both horizontal axis and vertical axis wind turbine blades. With the steadily increasing size of wind turbines, fatigue loads are become increasingly difficult to manage and this has motivated an approach called “smart rotor control” where active flow control devices and sensors are installed on the blades (eg Barlas and van Kuik, 2010). A major contributor to fatigue loading is the phenomenon of dynamic stall (eg Schreck & Robinson, 2005), which occurs when blades pitch rapidly beyond their static stall angle (αs). On the one hand, blades experience cyclically varying wind speeds and angles-of-attack depending upon their azimuthal location. On the other hand, stochastic changes in wind speed, in particular sudden gusts, cause the blades to experience an effective sudden pitch-up with a simultaneous increase in wind speed. Recently, we evaluated the ability of a modified Goman-Khrabrov (GK) model to predict the lift coefficient history during pitching maneuvers on a relatively thick NACA 0018 airfoil (Williams et al., 2015). It is a state space model that consists of a state equation for the internal dynamic variable (X) and an output equation for the lift coefficient. It requires two time constants and a map of the static internal variable X0 (α) determined from quasisteady experimental data. According to Goman-Khrabrov (1994) the function X0 nominally represents the degree of flow attachment over an airfoil during static conditions. Fully attached flow corresponds to X0= 1, and fully separated flow is indicated by X0= 0.

The existence of static hysteresis has a strong effect on the lift response during periodic pitching. In our earlier work the effects of hysteresis were modelled by means of a bifurcated X0 function, together with two additional constants determined by critical angles of attack of the quasi-steady lift curve. This produced a more general model that was capable of capturing the dynamic hysteresis for the thicker airfoil in light and deep stall conditions. Physically, this corresponded to modelling of dynamic trailing-edge stall. However, the modified model was not able to predict the effect of the dynamic stall vortex (DSV) during deep stall. Fortunately, flow control applied by means of steady blowing at both low and high momentum coefficients Cµ can totally eliminate the effect of the dynamic stall vortex. This raises the interesting and potentially very useful possibility that a modified form of the GK model can be capable of modeling the controlled cases where predominantly trailing-edge separation is present. The main objectives of this paper are to evaluate the ability of time-delay models to predict the lift coefficient history during pitching of the NACA 0018 when subjected to slot-blowing flow control, and to demonstrate dynamic lift control with a Simulink® simulation. If successful, this will have far-reaching implications for wind turbine blade dynamic stall control.