E-045

Three-way Catalytic Converter Modelling: A Machine Learning Approach for the Reaction Kinetics

Authors: Luigi Glielmo, Stefania Santini*, Michele Milano**, Gabriele Serra***
Affiliation: *Dipartimento di Informatica e Sistemistica, Università di Napoli Federico II, Napoli, Italy
**Institute of Fluid Dynamics, Swiss Federal Institute of Technology (ETH), Zurich, Switzerland
***Magneti Marelli Divisione Controllo Motore, Bologna, Italy

Abstract
In this paper we present a novel approach to the problem of three-way catalytic converter dynamic modeling; one of the main issues related to modeling these devices is the reaction kinetics submodel, that has to be at the same time simple and flexible enough to capture all the significant features of the real system. We propose the use of machine learning techniques to solve this problem: a neural network structure for the kinetic submodel and a genetic algorithm to tune its parameters. In this way the difficulties arising from the identification of the resulting overall model are avoided.

Luigi Glielmo, Stefania Santini
Dipartimento di Informatica e Sistemistica,
Università di Napoli Federico II
via Claudio 21, 80125 Napoli, Italy
Tel. +39-81-7683172
fax: +39-81-7683186
glielmo@disna.dis.unina.it
santini@disna.dis.unina.it

Michele Milano
Institute of Fluid Dynamics,
Swiss Federal Institute of Technology (ETH)
Sonneggstrasse 3, 8092 Zurich, Switzerland
milano@ifd.mavt.ethz.ch

Gabriele Serra
Magneti Marelli Divisione Controllo Motore
via Timavo 3, Bologna, Italy