Prediction of effect of tungsten filled Co-30Cr-4Mo-1Ni metal matrix biomedical composite alloy on sliding wear peculiarity using Taguchi methodology and ANN

Abstract

Artificial neural networks have appeared as a better candidate to arithmetical wear models, due to their competence of handling non linear behaviour, learning from experimental results and generalisation. In this study, an ANN technique was applied to predict the effect of tungsten filled particulates on sliding wear performance of fabricated Co-30Cr-4Mo-1Ni biomedical metal matrix alloy composite for hip implant application with distilled water medium. In order to appraise the behaviour of fabricated biomedical alloy composite fulfilling diversified performance measures, Taguchi methodology has been espoused. An orthogonal array and statistical analysis of variance were used to identify the significant factor setting for obtaining better performance output. Confirmation test were carried out to verify the experimental results. The surface morphology of the worn out surfaces and cross-sectional microstructure of the fabricated alloy composite were analysed by using SEM to understand the wear mechanism and microstructure. Finally, the responses have been predicted using both ANN and Taguchi method so that a comparative evaluation can be made. From this analysis, it can say that neural network predicts the responses more precisely than Taguchi prediction. This study will give an idea for hip implant application but not direct replacement of human joints.