Hybrid fuzzy control of linear and nonlinear systems
Ya Lei Sun, Meng Joo Er · 2002
A hybrid fuzzy controller suitable for controlling both linear and nonlinear systems is proposed. The proposed controller, comprising a linear proportional integral derivative (PID) controller and a linear fuzzy logic controller, employs genetic algorithms to facilitate optimal tuning of the controller gains. A two-input dynamic linear fuzzy logic controller with linearly defined fuzzy space is developed to replace the conventional PI controller in the PID connective structure. Closed-form analysis shows that the proposed fuzzy logic controller is capable of generating nonlinear output by using varying gains and dynamic fuzzy rule base. Simulation results for a direct-current motor and a tactical missile model demonstrate that the proposed controller outperforms other existing controllers, is robust and has great potential in many other industrial applications.