On-line control of nonlinear systems using a novel type-2 fuzzy logic method

Abstract

In this paper, a new type-2 fuzzy logic controller (T2FLC) is proposed for on-line control of nonlinear systems. The proposed method provides more flexibility in applying human thinking in control system design in comparison with the traditional or even type-1 fuzzy logic controllers. One encounters some problems using the type-2 fuzzy logic controllers on-line because of the large volume of computations required. The main aim of this paper is to propose a new type-2 fuzzy controller with a small computational burden so that it can be used on-line. In the proposed T2FLC, the programs related to the computation of union of all qualified consequent T2MFs are divided into some sub-problems, and to design the T2FLC -not interval -we merge the Type Reducer and Defuzzifier into one step. By doing so, the computational time reduces, and the performance of our T2FLC increases so that it can be used on-line in dynamic systems. We study the noise concealing properties of the proposed T2FLC as well. The proposed method is applied to two non-linear dynamic systems, ball & beam and inverted pendulum. The results are compared with PID and type-1 fuzzy logic controllers. In the simulation results, it is also shown that the proposed controller has good noise cancelling properties.