Answers are grounded in this paper and cite their sources.
Experimental study demonstrates that infill pattern dictates tensile strength and hardness in 3D-printed PLA, highlighting the utility of neural network modeling for property prediction.
This study experimentally and statistically examined the effects of infill pattern, printing temperature, and annealing duration on the mechanical properties of PLA tensile samples produced using the fused filament fabrication method (FFF). The samples were produced in three different infill patterns (grid, zigzag, and honeycomb) and at two different printing temperatures (200 and 230 ℃). The samples were annealed for 90 and 180 minutes according to the experimental design created using the DoE (Design of Experiment) approach. The surface hardness of the samples was measured and tensile tests were performed. The results were analyzed using Taguchi and ANOVA methods, so that the values and parameters effective on the results and obtaining the optimum results were determined. Additionally, a prediction model was created using the Artificial Neural Network method (ANN). Control samples were produced under different manufacturing parameters and tested. The accuracy of the model was tested by comparing the experimental results to the predicted values. The results indicated that the highest tensile strength was achieved with sample H-230-90 at 45.33 MPa. The honeycomb achieves the highest tensile strength and hardness values due to its homogeneous stress distribution, while zigzag patterned samples exhibit the highest ductility. It was found that the annealing process increased tensile strength and surface hardness, but decreased elongation. According to the ANOVA results, the most effective parameter on tensile strength was the infill pattern with 96.06%. The model yielded successful results in predicting grid and zigzag infill pattern samples but had high error rates in predicting honeycomb samples.
No takes yet. Share an insight, caveat, or question.
İsmail Aykut Karamanlı (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: