Recupero Password

Path: Home » Indice Pubblicazioni » Convegni ATI - Accesso riservato soci CTI » CA - 59 - Genova 2004 » Experimental identification of a recu...

Convegni ATI - Accesso riservato soci CTI

Experimental identification of a recurrent neural network for air-fuel ratio model-based control


Autore: I. Arsie, M. Marotta, C. Pianese, M. Sorrentino

Collana: CA - 59 - Genova 2004

The development of a Recurrent Neural Netwok (RNN) for the simulation of the wall wetting dynamics in SI engines is presented. A Multi Input-Single Output structure has been adopted, considering injected fuel, manifold pressure and engine speed as external input variables and the Air-Fuel Ratio at the exhaust gas oxygen sensor location as output. Furthermore, due to the nonlinear mapping capabilities and adaptive features of the RNNs, the use of the developed model in the framework of neural control architectures derived from the literature is discussed. The RNN has been trained fitting a set of transient data measured on a commercial 4 cylinders SI engine. The results show that the RNN model guarantees good accuracy and generalization thus making the RNN a valid candidate for advanced model based AFR control application.

Per scaricare il Documento devi essere loggato!

Via Scarlatti 29, 20124 Milano
Tel. +39 02 266.265.1
Fax +39 02 266.265.50
P.IVA 11494010157

Social CTI


Twitter LinkedIn

Disclaimer - Privacy