Abstract: Taking the NREL 5 MW reference wind turbine with trailing edge flaps as an object of study, an improved aeroelastic model was established for the smart blade considering the unsteady aerodynamics, aerodynamic damping and bend-twist coupling, of which the simulation results were compared with that of FAST platform. Based on the adaptive PID of RBF neural network, an active controller was designed for the trailing edge flap to control the deflection of blade tips under standard turbulent wind conditions. Results show that the accuracy of the improved aeroelastic model is relatively high; the active controller for the trailing edge flap can effectively reduce the fluctuation of the blade tip deflection.