EXPERIMENTAL INVESTIGATIONS ON THE EFFECT OF PROCESS PARAMETERS OF THE 3D-PRINTING USING FUSED DEPOSITION MODELLING

Abstract

This research work aims to model and predict Flexural strength and 0.2% offset yield strength of an acrylonitrile-butadiene-styrene (ABS) manufactured through 3D printing using FDM. This process been studied and optimised using a machine learning process. An artificial neural network (ANN) multi-parameter regression model was created and then input–output relations model developed by this network was optimised by genetic algorithm (GA) to avail optimised parameters. an L27 Taguchi array was used to combine the specific parameters (nozzle diameter, layer height, fill density, printing velocity, raster orientation and infill pattern) to perform experiments. All samples were subjected to a three-point bending test performed according to the ASTM D7264 standard aimed to obtain Flexural strength and the 0.2% offset yield strength. ANN models were highly optimized and provided a better approach for the prediction of higher values of E and RP0.2 than their experimental. ANN-GA based modelling and optimization suggest a direct relation between choosing process parameters correctly and enhancing 3D- printing performance. ANN GA approach provides a set of optimal solutions for obtaining suitable output values. Infill pattern plays a critical role in providing the strength. Patterns intermediated to linear, rectilinear and honeycomb can be suggested for desired Flexural strength and 0.2% offset yield strength within constraints of input factors.