Journal of Product Design, Quality Engineering & Technology

Authors: Rajendra M. Galagali, R. G. Tikotkar

Abstract: Austempered Ductile Iron (ADI) is emerging as an alternative material to forged steel in many applications. ADI possesses remarkable wear resistance properties with reasonable toughness and ductility combined in one. In the present investigation, ADI was obtained by solutionizing ductile iron at 870°C for 90 minutes and austempering at 345°C for 150 minutes. The present work aims at predicting the wear behaviour of ADI by using Artificial Neural Network (ANN). A feed forward back propagation algorithm was used for the analysis. Experiments were planned using Design of experiments by Taguchi method. L9 orthogonal array was selected and experiments were conducted as per the run order generated by Taguchi for the control factors at three levels. Applied load of 19.62, 29.43, 39.24N were selected for study at speeds of 1.047, 2.095, 3.142m/s for sliding distances of 1257, 2514 and 3770m. Finally well optimized and trained neural network with LM training algorithm is used to predict the wear properties as a function of testing conditions, according to the input data sets. The results show that the predicted data are perfectly acceptable when compared to the actual experimental results. Hence a well trained artificial neural network system is expected to be very useful for estimating the weight loss of ADI under dry sliding wear test conditions at atmospheric temperature.