JFET_V3_N2_RP9 D. Hanumantha Rao G.R.N. Tagore G. Ranga Janardhana Kim S. Kun Journal on Future Engineering and Technology 2230 – 7184 3 2 55 62 Secondary dendrite arm spacing, Artificial neural network, Genetic algorithm, Solidification simulations The chemical, mechanical, and physical properties of cast metal products are greatly determined by the microstructure formed during solidification. Microstructural parameter Secondary Dendrite Arm Spacing (SDAS) has been observed to have a significant impact upon the yield strength, ultimate tensile strength and elongation of cast products. A comprehensive model which can predict the SDAS values, will allow the aluminum industry to effectively troubleshoot, develop and improve different alloys. In the present investigation, a hybrid Artificial Neural Network (ANN) — Genetic Algorithm (GA) model is developed for predicting the SDAS values in aluminum alloy castings. Adaptation and optimization of network weights using GA is proposed as a mechanism to improve the performance of ANN model. The process parameters considered for predicting SDAS values are: chill volume heat capacity, insulation on riser and pouring temperature. Solidification simulations are carried out using finite difference method for obtaining the data in order to develop the ANN model. The proposed model is validated experimentally. November 2007 - January 2008 Copyright © 2008 i-manager publications. All rights reserved. i-manager Publications http://www.imanagerpublications.com/Article.aspx?ArticleId=678