An Investigation on Impact of Machine Learning in Additive Manufacturing

Thakar, Chetan M.; Ventayen, Randy Joy Magno; Ramírez-Asís, Edwin; Vilchez-Carcamo, Juan; Palma, Misael Erikson Maguiña; Thommandru, Abhishek

doi:10.1007/978-981-19-0108-9_32

Cited by 1 publication

(1 citation statement)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…29 ML methods can forecast future results by looking at historical and experimental data. 30 Findings indicated that the XGBoost model predicted melt pool width and depth with 97% and 96.3% accuracy, respectively. 31 In-depth analyses of ML-enhanced AM technology's advantages and potential applications are well-described and helpful for a special AM issue.…”

Section: Introductionmentioning

confidence: 94%

Hybrid additive manufacturing of highly sustainable Polylactic acid -Carbon Fiber-Polylactic acid sandwiched composite structures: Optimization and machine learning

Thakur

Kumar

Singh

et al. 2023

Journal of Thermoplastic Composite Materials

View full text Add to dashboard Cite

Carbon fibre (CF) based polymeric composites are being used in automobile and aviation applications due to their lightweight, excellent mechanical and physical properties. In this study, the fused filament fabrication (FFF) technique was used to prepare composite structures of polylactic acid (PLA) sandwiched with CF layers followed by prediction of optimum setting by machine learning (ML). In the first stage, PLA-CF-PLA based composite structures (as per ASTM D638 type IV) were manufactured with deposition of fibre at various angles (0°, 45° and 90°), nozzle temperature (200°C, 205°C and 210°C) and bed temperature (55°C, 60°C and 65°C). Further, the prepared composite structures were subjected to tensile testing (strength at peak and break, strain at peak and break, Young’s modulus and modulus of toughness) followed by fracture analysis through a scanning electron microscope (SEM) energy-dispersive spectroscopy (EDS). The results of the study are supported by X-ray diffraction (XRD), Fourier transforms infrared spectroscopy (FTIR) analysis, and differential scanning calorimetry (DSC). In the second stage, Classification and Regression Trees (CART) of the ML approach were used to model strength at peak and strength at break. The results of the study have highlighted that combination of parameters, 0˚orientation of CF deposition, 205°C nozzle temperature, and 55°C bed temperature are the optimum settings for manufacturing PLA-CF composite structures. The ML CART model is a valuable tool for predicting the strength at peak and strength at break hybrid additive manufacturing of highly sustainable PLA-CF-PLA sandwiched composite structures.

show abstract

Section: Introductionmentioning

confidence: 94%