Adaptive neuro-fuzzy controller for static VAR compensator to damp out wind energy conversion system oscillation

Wind shear and tower shadow produce a periodic pulse reduction in mechanical torque captured from wind energy resulting in wind energy conversion system (WECS) active power oscillations. In this study, an adaptive neuro-fuzzy controller for static VAR compensator, used in power networks integrated with WECS, is presented to address the torque oscillation problem. The proposed controller consists of a radial basis function neural network representing a third-order auto-regressive and moving average system model and performing the prediction, and a main controller with adaptive neuro-fuzzy inference system providing the damping signal. A modified two-area four-machine power network with WECS integration is applied to validate the proposed implementation, compared with conventional lead/lag compensation. Time-domain simulations prove that the proposed controller can provide a damping signal to improve the active power oscillation and system dynamic stability, influenced by torque oscillations under WECSs synchronised operating condition.