Evaluation of mechanical properties of fiber-reinforced syntactic foam thermoset composites: A robust artificial intelligence modeling approach for improved accuracy with little datasets

Abstract

Fiber accumulation due to printing ink inconsistency makes additive manufacturing (AM) of reinforced thermoset syntactic foam composites difficult. This study predicts and analyzes the mechanical properties of AM-made carbon fiber-reinforced syntactic thermoset composites to overcome experimental limitations. Thus, an adaptive neuro-fuzzy inference system (ANFIS)-based model creates an accurate mechanical behavior prediction under a variety of conditions without experimental inquiry. Compression and flexure tests assessed the ANFIS model's validation. The model's predictions were very close to reality, validating the approach taken to improve the technical assessment of the created composites, which are perfect for weight reduction, mechanical improvement, and product complexity.