Artificial neural-network based modeling of variable valve-timing in a spark-ignition engine Mustafa Go ¨lcu ¨ a, * , Yakup Sekmen a , Perihan Erduranlı b , M. Sahir Salman c a Department of Mechanical Education, Technical Education Faculty, Pamukkale University, 20017 Kinikli, Denizli, Turkey b Department of Mechanical Education, Technical Education Faculty, Zonguldak Karaelmas University, 78100 Karabuk, Turkey c Department of Mechanical Education, Technical Education Faculty, Gazi University, 06500 Teknikokullar, Ankara, Turkey Accepted 13 July 2004 Available online 5 January 2005 Abstract Variable valve-timing and lift are significant operating and design parameters affecting the performance and emissions in spark-ignition (SI) engines. Previous investigations have dem- onstrated that improvements in engine performance can be accomplished if the valve timing is variable. Traditionally, valve timing has been designed to optimize operation at high engine-speed and wide-open throttle conditions. Controlling valve timing can improve the tor- que and power curve of a given engine. Variable valve-timing can be used to reduce fuel con- sumption and increase engine performance. Intake valve-opening timing was changed from 10° crankshaft angle (CA) to 30° CA for both advance and retard with 10° CA intervals to the original opening timing. In this study, artificial neural-networks (ANNs) are used to deter- mine the effects of intake valve timing on the engine performance and fuel economy. Experi- mental studies were completed to obtain training and test data. Intake valve-timing and engine speed have been used as the input layer; engine torque and fuel consumption have been used as the output layer. For the torque testing data, root mean squared-error (RMSE), fraction of variance (R 2 ) and mean absolute percentage error (MAPE) were found to be 0.9017%, 0306-2619/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.apenergy.2004.07.008 * Corresponding author. Tel.: +90 258 213 4030/1528; fax: +90 258 211 0098. E-mail address: mgolcu@pamukkale.edu.tr (M. Go ¨lcu ¨). Applied Energy 81 (2005) 187–197 www.elsevier.com/locate/apenergy APPLIED ENERGY