Enhancing drilling performance in 3D printed PLA implants application of PIV and ML models

Area of Science:

• Additive Manufacturing
• Materials Science
• Mechanical Engineering

Background:

• Fused Deposition Modeling (FDM) enables rapid prototyping of complex geometries using biocompatible Polylactic Acid (PLA).
• Machining processes like drilling are crucial for achieving desired functional characteristics in FDM parts.
• Optimizing drilling parameters is essential for enhancing material removal rate (MRR) and hole quality.

Purpose of the Study:

• To investigate the effects of drilling parameters (spindle speed, feed rate, drill diameter) on FDM-PLA parts.
• To assess material removal rate (MRR), surface roughness (Ra, Rz), circularity, and cylindricity.
• To utilize the Proximity Indexed Value (PIV) tool and machine learning models for process optimization.

Main Methods:

• Fabrication of PLA parts using FDM.
• Drilling experiments with varying spindle speed, feed rate, and drill diameter.
• Application of Proximity Indexed Value (PIV) for optimization.
• Development and comparison of Artificial Neural Network (ANN), Support Vector Machine (SVM), and Random Forest (RF) models.

Main Results:

• The Random Forest (RF) model achieved a high R² value of 99.16%, outperforming ANN (89.45%) and SVM (89.08%).
• PIV tool effectively predicted near-optimal drilling parameters.
• Machine learning models demonstrated strong predictive capabilities for optimizing drilling outcomes.

Conclusions:

• Machine learning models, particularly Random Forest, provide accurate predictions for optimizing drilling parameters in FDM-PLA parts.
• The study successfully identified optimal drilling conditions to enhance MRR and hole quality.
• This research contributes to improved post-processing techniques for additive manufactured components.