Modified descend curvature based fixed form fuzzy optimal control of nonlinear dynamical systems

Abstract

In this study, a fuzzy rule-based optimal controller is designed for nonlinear dynamical systems. The direct second order method (or direct-descend-curvature algorithm) with a modification called "modified descend controller (MDC)" is used for calculating the parameters of the fuzzy feedback controller. The optimal control problem defined here has dynamic constraints of nonlinear system states and static constraint of a known form of fuzzy controller. The form used here is a standard fuzzy system that uses singleton fuzzifier, product inference engine, center average defuzzifier, and with the Gaussian membership functions of the system states to be controlled. The design is developed by minimizing a quadratic performance index selected for the desired operating conditions. Successful simulation results of controlling the temperature of a continuous stirred tank reactor (CSTR) and a bioreactor are given.