Optimization of Mechanical Properties of Cordia dichotoma/Alumina Nanoparticle Reinforced Epoxy Nanocomposites Under Alkali Treatment Using ANN Techniques

Abstract

The growing interest in natural fiber- and nano-filler-reinforced composites is driven by their excellent strength-to-weight ratio, recyclability, and environmental sustainability. This study aims to optimize key parameters influencing the mechanical performance of hybrid nanocomposites reinforced with Cordia dichotoma (CD) fibers and nano alumina (nAl2O3). The composites were fabricated using the conventional hand lay-up technique by varying the content of CD fiber (wt%), fiber length (mm), nAl2O3 content (wt%), and CD fiber alkali treatment duration. ASTM standards assess mechanical properties. The Taguchi optimization method was employed to identify optimal process parameters. Maximum tensile strength was achieved with 10 wt% Cordia dichotoma fiber (10 mm), 4 wt% nAl2O3, and 12 hours of NaOH treatment. Optimal flexural strength was obtained under similar conditions, with an 8-hour treatment, while the best impact strength was observed with a 15 mm fiber length, 6 wt% nAl2O3, and 12-hour alkali treatment. Experimental results closely aligned with artificial neural network (ANN) predictions, with deviations within 3-4%, confirming the model's predictive accuracy. ANOVA analysis revealed that fiber content, filler content, and alkali treatment contributed 4%, 1%, and 0.1%, respectively, to the variation in properties, with fiber content yielding approximately 90% influence. The close agreement between experimental data and ANN forecasts highlights the potential of ANN-based models for predicting mechanical behavior and optimizing composite design. This approach provides valuable insights for enhancing reliability and predicting service life in various industrial applications.

Topics & Concepts