Study on compressive strength characteristics of selective inhibition sintered UHMWPE specimens based on ANN and RSM approach

Abstract

Selective inhibition sintering (SIS) is an evolving additive manufacturing (AM) technology that enables to build functional parts from various polymers, metals and ceramics materials in reasonable cost and time. SIS parts are significantly influenced through their process parameters. Further, due to the brittleness and anisotropic characteristics of functional parts, optimization of SIS process variables is necessary to improve the compressive loading capacity. In this study, the effect of five main SIS process parameters such as infrared (IR) heater power, bed temperature, IR heater feedrate, layer thickness and IR heater scanning distance on the compressive strength (CS) of selective inhibition sintered ultra-high molecular weight poly ethylene (UHMWPE) test specimens was examined. Firstly, experimentations were conducted based on Box–Behnken design (BBD) scheme by considering three levels for each process parameters. Subsequently, the CS of SIS test specimens were predicted using response surface methodology (RSM) based on developed quadratic regression model. Finally, the predicted CS using RSM model were compared with that of artificial neural network (ANN) model. Further, the influence of SIS parameters on CS was studied through performing analysis of variance (ANOVA) and sensitivity analysis. The results show that the power and feed rate of the infrared heater have a significant influence on CS. Furthermore, an optimization was performed using the desirability approach to identify the optimal levels of the SIS process parameters and was validated through experimental result.