DBFNN based adaptive excitation controller of a power system using backstepping design

DBFNN (Direction Basis Neural Network) proposed recently had many novel properties. In this paper, a DBFNN based direct adaptive controller was designed for SISO strict-feedback system by using backstepping method. A virtual controller was designed in every step of backstepping and the real controller was acquired in the last step. The tuning law of NN weights was derived from a selected integral Lyapunov function. So the stability of the closed loop and convergence of weights were guaranteed. The proposed scheme was applied to design an excitation controller for a power system. The simulation demonstrates good tracking performance and robustness of the designed controller.