Predictive modeling of the mechanical behavior of 3D-printed polylactic acid/wood composite: Comparison of GEP and ANN methods

This study introduces a novel approach for predicting the mechanical properties of 3D-printed polylactic acid wood composites using gene expression programming (GEP) and artificial neural networks (ANN) modeling methods. Addressing the challenge of determining optimal process parameters in fused dep...

Full description

Saved in:
Bibliographic Details
Main Authors: Abhijit Bhowmik, Raman Kumar, Ranganathaswamy M. K., Y. Karun Kumar, Priyaranjan Samal, Abinash Mahapatro, Abdulaziz N. Alhazaa, Valentin Romanovski, A. Johnson Santhosh
Format: Article
Language:English
Published: AIP Publishing LLC 2025-04-01
Series:AIP Advances
Online Access:http://dx.doi.org/10.1063/5.0268653
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This study introduces a novel approach for predicting the mechanical properties of 3D-printed polylactic acid wood composites using gene expression programming (GEP) and artificial neural networks (ANN) modeling methods. Addressing the challenge of determining optimal process parameters in fused deposition modeling of natural fiber composites, experiments were designed using Taguchi’s L27 orthogonal array. Five key parameters were analyzed: layer thickness (100–300 μm), printing speed (40–90 mm/s), raster angle (0°–90°), infill density (35%–95%), and nozzle temperature (200–220 °C). ANOVA results identified raster angle as the most influential factor, contributing 38.36% and 26% to tensile and compressive strengths, respectively. Subsequently, a comparative statistical analysis evaluated the predictive accuracy of GEP and ANN. The GEP model exhibited superior performance, achieving validation errors between 0.04% and 0.82%, outperforming ANN (0.34%–5.31%). These findings provide a robust framework for enhancing the mechanical performance of sustainable 3D-printed composites, enabling more efficient and reliable production processes in additive manufacturing.
ISSN:2158-3226