Advancements of machine learning techniques in fiber-filled polymer composites: a review

Abstract

The integration of machine learning (ML) techniques in the characterization and optimization of fiber-filled polymer composites is a topic of increasing importance in industries such as aerospace, automotive, and construction. Traditional experimental methods for characterizing these composites can be time-consuming and limited in scope, driving the adoption of ML approaches. This review article explores various ML paradigms and their applications in polymer composite manufacturing and process simulation. The objective of the study is to investigate ML-based methods for predicting mechanical properties, optimizing fabrication processes, conducting microstructure analysis, and predictive modeling of composite performance. Furthermore, the review addresses challenges and identifies future research opportunities in leveraging ML for advancing composite material design and optimization. By synthesizing current research findings and highlighting potential areas for development, this review contributes to the ongoing exploration of ML’s role in revolutionizing the field of fiber-filled polymer composites.