Performance analysis of Lyapunov stability-based and ANFIS-based MRAC

Abstract

Analysis of two adaptive controller parameter adjustment laws for a model reference adaptive controller has been discussed in this paper. The comparison has been done about applying Lyapunov stability rule and using adaptive neuro fuzzy inference system (ANFIS) to adjust parameter for model reference adaptive control. Discussion of the nature of system, adaptive controller, basic block diagram and control law has been presented. For intense analysis two case studies have been considered. Simulation of two bench-mark process control applications, level control in coupled tank and concentration control in biochemical reactor (BCR) has been discussed. Comparative results have been plotted and discussed for each proposed algorithm. Considered systems have mutual parameter interaction and nonlinear parameter dynamics. Introduction part discusses literature survey, development of the topic and importance of the work. Initially Lyapunov rule-based technique has been applied for control in both cases. With ANFIS-based algorithm, new values of adjustment parameter have been generated. Results shows that performance of ANFIS-based model reference adaptive control (MRAC) gives improved results in presence of system uncertainties.