Selective laser sintering of carbon fiber–reinforced PA12: Gaussian process modeling and stochastic optimization of process variables

Abstract

This work presents an in-depth investigation of the influence of the individual laser sintering parameters on density, mechanical, and dimensional properties of carbon fiber–reinforced PA12 parts manufactured by selective laser sintering (SLS) . A space-filling design of experiments method was used to plan the experiments and SLS trials were conducted to manufacture test samples that were characterized in terms of dimensional accuracy, density, and mechanical properties. Gaussian process–supervised learning was used to model the interaction between laser sintering parameters and quality properties. Stochastic optimization via evolutionary algorithm was employed to obtain trade-off solutions for several multi-objective optimization tasks. The Gaussian process presented excellent model quality for the majority of response variables evaluated. Laser sintering parameters had a significant influence on physical and mechanical properties, exhibiting complex and non-linear behavior. Multi-objective optimization showed a wide range of optimized laser sintering parameters available, depending on the trade-off objective desired.