INTELLIGENT FUZZY IMPROVEMENT OF NON-SINGULAR TERMINAL SLIDING-MODE CONTROL

Ali Esmaeili Jajarm, Amir H. Abolmasoumi, and Hamid R. Momeni

Keywords

ERWLS training, non-singular terminal sliding mode control, T–S fuzzy inference

Abstract

This paper presents a novel hybrid method for non-linear control using non-singular terminal sliding mode control (NTSMC) together with a fuzzy inference system. This work obtains the sliding surface exponents to decrease both convergence time and output tracking error. The NTSMC method is applied to deal with the uncertainties in system and to improve the finite-time convergence of states to equilibrium point. To determine error exponents in non-linear sliding surface optimally, Takagi–Sugeno (T–S) fuzzy inference system is suggested. The fuzzy rules are trained with extended recursive weighted least squares (ERWLS) algorithm. The stability of defined sliding surface is proved by Lyapunov theory. The effectiveness of the proposed approach is demonstrated by applying it on a magnetic bearing system.

Important Links:

Go Back