Simulation of a fuzzy-PD learning control system

Conventional PD-type controllers have poor performances when dealing with nonlinear plants outside the linear domain for which they have been designed. In this paper we apply a learning algorithm to a fuzzy-PD controller to improve its performance in controlling a nonlinear plant. The previously designed fuzzy-PD control system is adapted by a learning mechanism which adjusts the controllerpsilas rule base so that the closed loop system behaves according to a reference model. The performances of the presented control structure, as applied to a nonlinear electromagnetic levitation system, are evaluated and shown by means of digital simulation.