Title: Adaptive neuro-fuzzy inference system in modelling damping performance of epoxy polymer concrete

Authors: Raman Bedi; Amritpreet Singh

Addresses: Department of Mechanical Engineering, National Institute of Technology, G.T. Road Bye Pass, Jalandhar-144011, India ' Department of Mechanical Engineering, National Institute of Technology, G.T. Road Bye Pass, Jalandhar-144011, India

Abstract: In this study the damping property of the epoxy polymer concrete is analysed in relation to its composition, i.e., percentage of epoxy and percentage of filler. Foundry sand having a mean particle size in 300 to 450 µm range is used as aggregate in polymer concrete. Twelve different compositions of polymer concrete are evaluated in this study considering the amount of epoxy resin and filler as variables. The testing of prepared samples is performed using dynamic mechanical analyser (DMA) technique. Damping of the prepared specimens has been evaluated at 10, 30 and 50 Hz. It is observed that resin percentage is most important factor affecting damping of epoxy polymer concrete. Adaptive neuro-fuzzy inference system (ANFIS) has been successfully used for modelling of damping behaviour of the epoxy polymer concrete. For all cases studied, an optimum selection of the training set for reliable modelling and elimination of the experimental cost was found to be between 80% and 70% of the available experimental data.

Keywords: epoxy polymer concrete; damping performance; adaptive neuro-fuzzy inference system; ANFIS; neural networks; fuzzy logic; modelling.

DOI: 10.1504/IJMATEI.2013.052794

International Journal of Materials Engineering Innovation, 2013 Vol.4 No.1, pp.18 - 34

Received: 05 Nov 2011
Accepted: 25 Apr 2012
Published online: 19 Jul 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article