<p>This study investigates the influence of raster angle and infill density on the elastic behavior of FDM-printed polylactic acid (PLA) parts, extending our prior raster-angle-only homogenization approach. Experimental tensile and compression tests were conducted on specimens with raster angles ranging from 0° to 90° and infill densities between 20% and 100%. Numerical homogenization was performed using representative volume elements (RVEs) derived from over 100 high-resolution optical microscopy images of real mesostructures and clustered via K-Means algorithms to capture authentic interlayer voids and filament morphologies. All nine orthotropic constants (E<sub>x</sub>, E<sub>y</sub>, E<sub>z</sub>, G<sub>xy</sub>, G<sub>yz</sub>, G<sub>xz</sub>, v<sub>xy</sub>, ν<sub>yz</sub>, and ν<sub>xz</sub>) are evaluated through homogenization; however, the combined dependence on raster angle and infill density is explicitly formulated only for the elastic moduli Ex, Ey, and Ez. These moduli exhibit quadratic dependencies on infill density, combined with angular averaging rules for consecutive-layer orientations. For the out-of-plane modulus, an empirical formulation incorporating a reference modulus at 45° and a harmonic averaging of layer contributions is introduced, motivated by enhanced diagonal load transfer observed in the homogenization results. A compact closed-form predictive law is proposed for the elastic moduli, enabling accurate estimation for arbitrary consecutive-layer orientations and infill densities, with relative errors consistently below 5% when validated against experimental data and detailed simulations (improved from 7% in our earlier angle-focused study). While the combined prediction of shear moduli and Poisson’s ratios is beyond the scope of this work and will be addressed in future studies, the proposed framework provides a reliable and reproducible tool for mechanical modeling, design optimization, material efficiency, and performance assessment in FDM additive manufacturing.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Influence of raster angle and infill density on orthotropic elastic properties of FDM-printed PLA: closed-form predictive modeling using K-Means clustering on real mesostructures for accurate RVE homogenization with experimental validation

  • Hamza Ait Benaissa,
  • Nabil Moujibi,
  • Hamid Zaghar

摘要

This study investigates the influence of raster angle and infill density on the elastic behavior of FDM-printed polylactic acid (PLA) parts, extending our prior raster-angle-only homogenization approach. Experimental tensile and compression tests were conducted on specimens with raster angles ranging from 0° to 90° and infill densities between 20% and 100%. Numerical homogenization was performed using representative volume elements (RVEs) derived from over 100 high-resolution optical microscopy images of real mesostructures and clustered via K-Means algorithms to capture authentic interlayer voids and filament morphologies. All nine orthotropic constants (Ex, Ey, Ez, Gxy, Gyz, Gxz, vxy, νyz, and νxz) are evaluated through homogenization; however, the combined dependence on raster angle and infill density is explicitly formulated only for the elastic moduli Ex, Ey, and Ez. These moduli exhibit quadratic dependencies on infill density, combined with angular averaging rules for consecutive-layer orientations. For the out-of-plane modulus, an empirical formulation incorporating a reference modulus at 45° and a harmonic averaging of layer contributions is introduced, motivated by enhanced diagonal load transfer observed in the homogenization results. A compact closed-form predictive law is proposed for the elastic moduli, enabling accurate estimation for arbitrary consecutive-layer orientations and infill densities, with relative errors consistently below 5% when validated against experimental data and detailed simulations (improved from 7% in our earlier angle-focused study). While the combined prediction of shear moduli and Poisson’s ratios is beyond the scope of this work and will be addressed in future studies, the proposed framework provides a reliable and reproducible tool for mechanical modeling, design optimization, material efficiency, and performance assessment in FDM additive manufacturing.