Experimental Study and ANN Development for Modeling Tensile and Surface Quality of Fiber-Reinforced Nylon Composites DOI Open Access
Osman Ülkir, Fatma Kuncan, Fatma Didem Alay

и другие.

Polymers, Год журнала: 2025, Номер 17(11), С. 1528 - 1528

Опубликована: Май 30, 2025

Additive manufacturing (AM) is gaining widespread adoption in the industry due to its capability fabricate intricate and high-performance components. In parallel, increasing emphasis on functional materials AM has highlighted critical need for accurate prediction of mechanical behavior composite systems. This study experimentally investigates tensile strength surface quality carbon fiber-reinforced nylon composites (PA12-CF) fabricated via fused deposition modeling (FDM) models their using artificial neural networks (ANNs). A Taguchi L27 orthogonal array was employed design experiments involving five printing parameters: layer thickness (100, 200, 300 µm), infill pattern (gyroid, honeycomb, triangles), nozzle temperature (250, 270, 290 °C), speed (50, 100, 150 mm/s), density (30, 60, 90%). An analysis variance (ANOVA) revealed that had most significant influence resulting strength, contributing 53.47% variation, with substantially at higher densities. contrast, dominant factor determining roughness, accounting 53.84% thinner layers yielding smoother surfaces. Mechanistically, a enhances internal structural integrity parts, leading an improved load-bearing capacity, while improve interlayer adhesion finish. The highest achieved 69.65 MPa, lowest roughness recorded 9.18 µm. ANN model developed predict both based input parameters. Its performance compared three other machine learning (ML) algorithms: support vector regression (SVR), random forest (RFR), XGBoost. exhibited superior predictive accuracy, coefficient determination (R2 > 0.9912) mean validation error below 0.41% outputs. These findings demonstrate effectiveness ANNs complex relationships between FDM parameters properties highlight potential integrating ML statistical optimize composites.

Язык: Английский

Mixed-mode dynamic crack propagation analysis in anisotropic functionally varying microcellular structures DOI Creative Commons

Victor Bautista,

Behnam Shahbazian,

MirMilad Mirsayar

и другие.

Results in Engineering, Год журнала: 2025, Номер unknown, С. 104117 - 104117

Опубликована: Янв. 1, 2025

Язык: Английский

Процитировано

0
Layer combination of similar infill patterns on the tensile and compression behavior of 3D printed PLA DOI Creative Commons

Menna G Aboelella,

Samy Ebeid,

Moustafa Mamoud Sayed

и другие.

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Апрель 6, 2025

Abstract With the growing popularity of 3D-printed products, material consumption has been a major concern in additive manufacturing recent years. Choosing infill structure and printing parameters for an application can be challenging product designers engineers, which lead to reduced increased cost savings while maintaining functioning. This study investigates mechanical behavior PLA structures by exploring influence multi-layer patterns on tensile compressive strength. Three common (triangular, grid, honeycomb) were evaluated at 20% 50% densities. A novel approach was employed, incorporating specimens with single-, two-, four-layer same pattern combinations, where subsequent layers rotated 180 degrees enhance interlayer bonding. Results demonstrated significant improvements both (up 64%) strength 47%) two-layer compared single-layer counterparts. The findings provide valuable insights into optimizing design layer configurations improved efficiency structures. research highlights potential part performance through strategic design, offering pathway toward enhanced properties manufacturing.

Язык: Английский

Процитировано

0
Experimental Study and ANN Development for Modeling Tensile and Surface Quality of Fiber-Reinforced Nylon Composites DOI Open Access
Osman Ülkir, Fatma Kuncan, Fatma Didem Alay

и другие.

Polymers, Год журнала: 2025, Номер 17(11), С. 1528 - 1528

Опубликована: Май 30, 2025

Additive manufacturing (AM) is gaining widespread adoption in the industry due to its capability fabricate intricate and high-performance components. In parallel, increasing emphasis on functional materials AM has highlighted critical need for accurate prediction of mechanical behavior composite systems. This study experimentally investigates tensile strength surface quality carbon fiber-reinforced nylon composites (PA12-CF) fabricated via fused deposition modeling (FDM) models their using artificial neural networks (ANNs). A Taguchi L27 orthogonal array was employed design experiments involving five printing parameters: layer thickness (100, 200, 300 µm), infill pattern (gyroid, honeycomb, triangles), nozzle temperature (250, 270, 290 °C), speed (50, 100, 150 mm/s), density (30, 60, 90%). An analysis variance (ANOVA) revealed that had most significant influence resulting strength, contributing 53.47% variation, with substantially at higher densities. contrast, dominant factor determining roughness, accounting 53.84% thinner layers yielding smoother surfaces. Mechanistically, a enhances internal structural integrity parts, leading an improved load-bearing capacity, while improve interlayer adhesion finish. The highest achieved 69.65 MPa, lowest roughness recorded 9.18 µm. ANN model developed predict both based input parameters. Its performance compared three other machine learning (ML) algorithms: support vector regression (SVR), random forest (RFR), XGBoost. exhibited superior predictive accuracy, coefficient determination (R2 > 0.9912) mean validation error below 0.41% outputs. These findings demonstrate effectiveness ANNs complex relationships between FDM parameters properties highlight potential integrating ML statistical optimize composites.

Язык: Английский

Процитировано

0