Skip to Main Content
Based on analyzing the effect of the scaling factors of a fuzzy controller on system performance, the paper proposes a new type of fuzzy controller with nonlinear scaling functions and discusses its effect on membership functions, equivalent mapping between input and output parameter, nonlinear surface for the controller and system performance in detail. Besides the parts mentioned above, the paper also studies the relationship of the parameters in nonlinear scaling functions and system performance. The parameter design of the fuzzy controller with nonlinear scaling functions is simple and normalization is also easy for online learning and adjustment in an adaptive fuzzy controller. Simulation result shows that compared with a conventional fuzzy controller, this controller with nonlinear scaling function has better dynamic and static properties, and it is a simple, useful and effective method for improving the fuzzy controller properties.
Machine Learning and Cybernetics, 2002. Proceedings. 2002 International Conference on (Volume:4 )
Date of Conference: 4-5 Nov. 2002