Effect of Infill Patterns with Machine Learning Techniques on the Tensile Properties of Polylactic Acid-Based Ceramic Materials with Fused Filament Fabrication

Abstract

The field of additive manufacturing is quickly evolving from prototyping to manufacturing. Researchers are looking for the best parameters to boost mechanical strength as the demand for three-dimensional (3D) printers grows. The goal of this research is to find the best infill pattern settings for a polylactic acid (PLA)-based ceramic material with a universal testing machine; the impact of significant printing considerations was investigated. An X-ray diffractometer and energy-dispersive X-ray spectroscopy with an attachment of scanning electron microscopy were used to investigate the crystalline structure and microstructure of PLA-based ceramic materials. Tensile testing of PLAbased ceramics using a dog bone specimen was printed with various patterns, as per ASTM D638-10. The cross pattern had a high strength of 16.944 MPa, while the tri-hexagon had a peak intensity of 16.108 MPa. Cross3D and cubic subdivisions have values of 4.802 and 4.803 MPa, respectively. Incorporating the machine learning concepts in this context is to predict the optimal infill pattern for robust strength and other mechanical properties of the PLA-based ceramic model. It helps to rally the precision and efficacy of the procedure by automating the job that would entail substantial physical effort. Implementing the machine learning technique to this work produced the output as cross and tri-hexagon are the efficient ones out of the 13 patterns compared.