Model-based optimal boundary control of selective catalytic reduction in diesel-powered vehicles. (November 2018)

• Record Type: Journal Article
• Authors: Aksikas, A.; Aksikas, I.; Hayes, R.E.; Forbes, J.F.
• Is Part Of: Journal of process control. Volume 71(2018)
• Page Start: 63
• Page End: 74
• Publication Date: 2018-11
• Subjects: Boundary system control -- Parabolic PDEs -- Hyperbolic PDEs -- Linear quadratic optimal control -- Selective catalytic reduction -- Diesel engines -- Distributed parameter systems; Process control -- Periodicals; Fabrication -- Contrôle -- Périodiques; Process control; Periodicals; Electronic journals; 660.281
• DOI: 10.1016/j.jprocont.2018.09.008
• Languages: English
• ISSNs: 0959-1524

Abstract

Highlights: The SCR dynamics and its mathematical representation have been described. An approximation technique for the computation of the eigenvalues and eigenvectors of the parabolic subsystem has been presented. Dynamical properties, including exponential stabilizability and exponential detectability, of the model are analyzed. Optimal linear-quadratic boundary control problem has been solved through the corresponding Riccati equation. A decoupling algorithm to solve it has been presented. Numerical simulations have been performed to evaluate the closed loop performance of the designed controller. Abstract: This paper is devoted to design a model-based boundary optimal controller for selective catalytic reduction system. The mathematical model consists of coupled parabolic-hyperbolic PDEs with an ODE. The main objective is to manipulate the ammonia gas at the inlet of the SCR in order to reduce the amount of NO x and ammonia slip as much as possible. The augmented infinite-dimensional state space representation has been used in order to solve the corresponding linear-quadratic control problem. The dynamical properties of both the linearized system and its augmented version have been studied. Under some technical conditions, it has been shown that the augmented system generates an exponentially stabilizable and detectable C 0 -semigroups. The linear-quadratic control problem has been solved for the augmented system. A decoupling technique has been implemented to decouple and solve the corresponding Riccati equation. An algorithm has been developed to describe the steps of solving the Riccati equation. Numerical simulations for the closed-loop system have been implemented to show the controller performances.