Affiliation:
1. Centre for Molecular Medicine and Diagnostics, Saveetha Dental College, Saveetha Institute of Medical and Technical Sciences, Saveetha University, Chennai, India
2. Department of Computer Science and Engineering, Saveetha Engineering College (Autonomous), Chennai, India
3. Department of Computer science, Chennai Institute of Technology, Chennai, India
4. Department of Dental Health, College of Applied Medical Sciences, King Saud University, Riyadh, Saudi Arabia
5. Restorative Dental sciences, Dental Materials Division, Faculty of Dentistry, King Saud University, Riyadh, Saudi Arabia
Abstract
The utilization of Fused Filament Fabrication (FFF) technology for developing sandwich structures proves to be an effective approach, enabling the rapid construction of intricate profiles and gaining widespread recognition for diverse structural applications. In this study, hexagon lattice-cored sandwich structures are created by situating the lattice core at the center of the PLA polymeric specimens. The performance is assessed by varying 3D-Printing Factors (3D-PFs), including Nozzle Temperature (NT), Layer Height (LH), Printing Speed (PS), and Line Width (LW). The levels of 3D-PFs are manipulated as follows: NT (180, 190, 200, 210°C), LH (0.15, 0.2, 0.25, 0.3 mm), PS (15, 20, 25, 30 mm/sec), and LW (0.1, 0.2, 0.3, 0.4 mm). By employing a FFF 3D printer, the sandwich specimens are 3D-printed and their compression properties are assessed using a Universal Testing Machine (UTM). In this research, various Machine Learning (ML) models namely Bayesian Ridge regression (BRid), Elastic Net linear regression (EN), Quantile Regression (QR), and Support Vector Machine (SVM) are utilized to predict the compressive strength/density property of the developed sandwich structure. This aids in determining the optimal levels of 3D-PFs to achieve enhanced compressive strength/density. The results reveal that the QR model, particularly when employed in the boosting ensemble technique, exhibits superior accuracy with a Root Mean Square Error (RMSE) of 0.26 × 104, Mean Absolute Error (MAE) of 0.21 × 104, and Median Absolute Error (MedAE) of 0.16 × 104. Utilizing the QR model within the boosting ensemble technique, the influence of 3D-PFs on resulting compressive strength/density is analyzed, facilitating the identification of optimized 3D-PF levels for improved compressive strength/density. Sandwich structures fabricated at these optimized levels demonstrate enhanced compressive properties, making them suitable for a variety of structural applications.
Funder
Researchers supporting scheme