Design of an Adaptive Controller with Online Identification Applied to a Pressure Tank
Dávalos-Pinto José, Acero-Coila Eloy

Abstract
The objective is to design an adaptive controller for a pressure tank using online identification to obtain the mathematical model of the system. The method considered nonparametric and parametric identification recursive least squares to obtain a ARMAX model of the system dynamics. An adaptive controller for controlling autotuning pressure was developed. The simulation was carried out by two operating ranges. The results show that it has reached a convergence time of 1 s., an identification with extended least squares, a settling time of 10 s. and steady-state error of 3% for the controlled variable. Based on the study, it is concluded that the controller has an acceptable performance against the nonlinearities of the process and does not require re-tune the controller to significant changes in the model or due to perturbance.

Full Text: PDF