Biniwale, Rajesh and Labhsetwar, Nitin and Kumar, R and Hasan, M Z (2002) Catalytic Converter Modeling: Artificial Neural Networks for Perovskite Based Catalyst. Society of Automotive Engineers. ISSN 0148-7191

PDF - Published Version Restricted to Registered users only Download (94Kb) | Request a copy

Abstract

Two-stroke vehicles including two and three wheelers constitute about 62% of vehicles in India, and about 45-50% in other developing countries in the region. These are the major contributors to vehicular pollution. Catalytic converters based on perovskite have been developed for 2- stroke vehicles. Detailed characterization was carried out during development of alumina washcoat and synthesis of perovskite catalyst to establish the thermal stability of alumina washcoat and phase formation of catalysts. A number of prototypes based on alumina-supported perovskite have been prepared and tested for mass conversion efficiency with respect to CO, HC and NOx using the Indian Driving Cycle (IDC). A catalytic converter model has been developed using the MATLAB artificial neural network toolbox for performance prediction. Experimental data generated during the detailed characterization of catalytic converters and its evaluation on engine dynamometer has been used as training data. The model was used for prediction of conversion efficiencies and mid-bed temperature. Keep-one-out method was used for comparison of predicted and experimental values. The algorithm developed predicts the performance very well and will be able to give prior information on the performance in view of future emission standards.

Item Type:	Article
Subjects:	Materials Science
Divisions:	UNSPECIFIED
Depositing User:	Dr. Rajesh Biniwale
Date Deposited:	12 Jan 2018 05:42
Last Modified:	12 Jan 2018 05:42
URI:	http://neeri.csircentral.net/id/eprint/1105

Actions (login required)

View Item