Analysis of Dry Sliding Wear Behaviour of Rice Husk Filled Epoxy Composites Using Design of Experiment and ANN

AbstractArtificial neural network (ANN) is a technique which can be used to simulate a wide variety of complex nonlinear engineering problems such as tribological performance of polymer composites. This article reports the implementation of ANN in analyzing the sliding wear performance of a new class of epoxy based composites filled with rice husk. Composites of four different compositions (5,10, 15 and 20wt.% of rice husk reinforced in epoxy resin) are prepared in simple hand-lay-up technique. Physical, chemical and mechanical tests are conducted on these composites. Dry sliding wear experiments are conducted as per Taguchi's orthogonal array design. Significant control factors affecting specific wear rate are identified. Based on the data obtained from experiments, an ANN model is trained and tested to predict the effect of wear behaviour on various control factors. Factors like sliding velocity, filler content and normal load, in this sequence, are the significant factors affecting the specific wear rate. This work shows that rice husk possesses good filler characteristics, as it improves the sliding wear resistance of the polymer resin