Prediction of tribological behaviour of rice husk ash reinforced aluminum alloy matrix composites using artificial neural network

Abstract

Artificial Neural Network (ANN) model was developed to predict the wear rate and coefficient of friction for the Rice Husk Ash (RHA) reinforced aluminum alloy composite. The composite was fabricated using the stir cast route and their tribological behavior was tested using Pin-on-Disc wear tester. The experiments were conducted based on Orthogonal array (L27) generated through the Taguchi Technique and their results were used to train the ANN model. The input parameters assigned to develop an ANN model are applied load, sliding speed, RHA particle size and weight percentage of RHA reinforcement. A four layer perception network having 4-7-8-2 architecture was found to be the optimum network. Finally, confirmation test was done to verify the predictive model with the experimental results and also the wear surface morphology of the wear pin was analyzed using the scanning electron microscope.