Nonlinear power control of doubly fed induction generator wind turbines using neural networks
A neural network controller is presented for the direct power control of a doubly fed induction generator (DFIG) in a wind turbine. Due to the variability of wind speed, aerodynamic control of the pitch angle of the turbine blades is necessary to capture maximum power from the wind. An alternative control strategy for tracking the maximum power that can be extracted is by accurate control of the turbine rotational speed. However, due to the nonlinearities inherent in a wind energy conversion system (WECS) and wind speed unpredictability, exact control of this speed becomes challenging. This is addressed by the use of an adaptive learning rate to guarantee convergence. Hence, the controller is robust to uncertainties in wind speed. The superior performance of the proposed control technique, under varying nominal wind speeds, is established by simulation results.
|Conference||25th IEEE International Symposium on Industrial Electronics, ISIE 2016|
Chaoui, H, & Okoye, O. (Okezie). (2016). Nonlinear power control of doubly fed induction generator wind turbines using neural networks. In IEEE International Symposium on Industrial Electronics (pp. 562–567). doi:10.1109/ISIE.2016.7744951