FGRN nonlinear controller and its applications

摘要

In this paper we propose a controller architecture based on our adaptive network called Fuzzy Graphic Rule Network (FGRN). In FGRN, the THEN part membership function and defuzzification steps are combined together, as a result the overall structure becomes simple. Moreover the initial setting of FGRN's parameters can be selected based on expert knowledge. FGRN's parameters can be adjusted using a method based on steepest descent algorithm and Lyapunov stability criteria. The performance of the proposed controller is represented by nonlinear plants, which are the water bath temperature and the High Voltage Direct Current (HVDC).