MULTI-MODEL ADAPTIVE CONTROL FOR NONLINEAR SYSTEMS

A multiple models adaptive control system will be presented. The advantages of this control strategy with respect the classical control will be illustrated on a level control system with nonlinear model plant. A recent recursive identification method, in open and closed-loop, and a R-S-T control algorithm have been proposed to guarantee the desired performances in the adaptive control scheme. Experimental results obtained from using this procedure prove that the real time implementation of this control mechanism is suitable for nonlinear processes with typical large parameter variation. The experiment is also available as a virtual laboratory for remote control.