In this study, an approach for thinking the three-body biting wear behavior of unfilled and element filled element fabric supported epoxy composite using two posing techniques – Taguchi study and artificial interconnected system are presented. A set of experiments were transported using an orthogonal array established Taguchi techniques to get data in a reserved manner. The results showed that the adding of graphite coarse into carbon gluing composite led to a decrease in its caustic wear resistance, and the wear deficit increased accompanying an increase in abrading distance and loads. To investigate the effect of control limits on the wear behavior of the composites, an reasoning of variance was acted, and the S/N ratio was planned. The results found that the normal load had the maximal physical in addition to statistical influence on the nasty wear of the composites followed by abrading distance and stuffing content. To predict the wear characteristics of composites as a function of testing environments, 3-[5]1-1 neural network architecture accompanying Levenberg Marquardt (LM) training invention was used. By equating the correlations obtained by Taguchi reversion analysis and pretended neural network accompanying the experimental results it was erect that the artificial neural network foresees the wear rate better than regression reasoning. Therefore, a well-trained fake neural network system maybe very helpful in judging the weight deficit in the complex three-body abrasive wear position of polymer composites.
Author(s) Details:
K. Sudarshan Rao,
Department of Mechanical Engineering, Shri Madhwa Vadiraja Institute of Technology and Management, Bantakal, Udupi, Karnataka, India.
Please see the link here: https://stm.bookpi.org/RPST-V7/article/view/9949
Keywords: Carbon fabric, epoxy, graphite filler, abrasive wear, Taguchi analysis, neural network
