Multi-Model Adaptive Control Systems

Abstract

A system with multiple models and adaptive control will be presented. The advantages of
this control with respect to the classical control will be illustrated on a level control system with
nonlinear model plant. A recent recursive method in open and closed-loop identification and an RS-
T controller design has been proposed to guarantee the performances in the adaptive control
scheme. The real time control system implementation confirms the opportunity of using the multimodels
adaptive control architecture in the case when the nonlinear plant model introduces a
typical large parameter variation. The article presents also an method to switch the algorithms.
The obtained results for tested instalation show that the control strategy based on a single model
and on a single controller generates a time response that is more affected by noise than the
response given by an adaptive strategy. The multiple models adaptive control procedure proposed
has the following advantages: a more precise model is chosen for the closed loop operating
system, the R-S-T adaptive control ensures very good real time results for closed loop nonlinear
system. It can be appreciate that the multiple models adaptive control can be recommended to
improve the performances of the nonlinear control systems.