A comparative study for the boundary control of a reaction-diffusion process: MPC vs backstepping

Abdelhamid Ouali (1) , Matthieu Fruchard (1) , Estelle Courtial (1) , Youssoufi Touré (1)

1 IRAUS - Département Images, Robotique, Automatique et Signal [Orléans]

In this paper, we consider a reaction-diffusion process described by a linear parabolic partial derivative equation (PDE). Two radically different control approaches are compared: the model predictive control (MPC) and the backstepping approach. The stabilization of the unstable reaction-diffusion process is first studied. Then, to deal with parameter uncertainties, an adaptive backstepping controller is developed and compared to a model predictive controller based on an internal model control (IMC) structure. Simulation results illustrate the efficiency of the two approaches in terms of precision and computational time.

Mots clés: Model Predictive Control; Backstepping control; Boundary control of PDEs

Domaines: Automatique / Robotique

Abdelhamid Ouali, Matthieu Fruchard, Estelle Courtial, Youssoufi Touré. A comparative study for the boundary control of a reaction-diffusion process: MPC vs backstepping. IFAC NMPC'12, Aug 2012, Noordwijkerhout, Netherlands. pp 200-206. ⟨hal-00655995v2⟩