Backstepping control of digital excitation systems based on neural network

Aimed at the characters of serious nonlinear and unknown or uncertain parameters, a backstepping control method based on neural networks is proposed to realize generator real-time control. Neural networks(NNs) are used to solve the contradiction of backstepping control and unmatching of systems. The real time simulation results show the control backstepping algorithm based on NNs has good properties, such as small overshoot, short tuning time and etc. And it is superior to the traditional PID strategy in terms of robustness and tracking ability. The method offers a new throughway for design and studying excitation control system.