Experimental comparison of new adaptive PI controllers based on the ultra-local model parameter identification

Hajer Thabet (1, 2) , Mounir Ayadi (1, 2) , Frédéric Rotella (3)

1 ENIT - Ecole Nationale d'Ingénieurs de Tunis
2 University of Tunis El Manar
3 LGP - Laboratoire Génie de Production

This paper is devoted to an experimental comparison between two different methods of ultra-local model control. The concept of the first proposed technique is based on the linear system resolution technique to estimate the ultra-local model parameters. The second proposed method is based on the linear adaptive observer which allows the joint estimation of state and unknown system parameters. The closed-loop control is implemented via an adaptive PID controller. In order to show the efficiency of these two control strategies, experimental validations are carried out on a two-tank system. The experimental results show the effectiveness and robustness of the proposed controllers.

Mots clés: Adaptive PID controller; Least squares method; Linear adaptive observer; Parameter estimation; Robustness; Trajectory tracking; Two-tank system; Ultra-local model control

Domaines: Automatique / Robotique

Hajer Thabet, Mounir Ayadi, Frédéric Rotella. Experimental comparison of new adaptive PI controllers based on the ultra-local model parameter identification. International Journal of Control, Automation and Systems, 2016, 14 (6), pp.1520-1527. ⟨10.1007/s12555-014-0550-1⟩. ⟨hal-02128148⟩