Analysis of wear assessment of FDM printed specimens with PLA, multi-material and ABS via hybrid algorithms

Abstract

The present work deals with wear assessment of FDM printed specimens with PLA (polylactic acid), Multi-Material, and ABS (acrylonitrile butadiene styrene). As per the American Society for Testing and Material G99 17 standard, the parts have been fabricated. The significant parameters raster angle, infill density, speed, temperature, wall thickness, and material selection have been taken as input factors to fabricate various specimens. The experimental matrix has been designed using optimal custom design for the response surface method, accommodating custom model, categoric elements, and irregular (constrained) regions. A total of 38 experimental runs have been achieved using the present approach. The 38 specimens have been printed with an FDM printer on these significant input factors and tested on the Pin on disc. An artificial neural network has developed a relationship between input factors and output response. GA-ANN has been applied to minimize the wear rate. It is observed that the minimum wear rate achieved 0.155371 mm3/m at PLA material at raster angle 89.258°, infill density 95.207%, temperature 220.009 ℃, speed 40.043 mm/s, and wall thickness 1.198 mm; this data has been validated using Hybrid algorithms.