Knowledge Discovery System For Fiber Reinforced Polymer Matrix Composite Laminate

TL;DR

This paper presents a knowledge discovery system for analyzing fiber orientations in polymer composites, using fuzzy classification and rule-of-mixture models to optimize material design for strength and weight.

Contribution

It introduces a novel knowledge discovery system integrating fuzzy classification and data mining for composite material analysis and design optimization.

Findings

Effective classification of fiber types based on length and orientation.

Graphical and statistical analysis of composite properties.

Support for designing lightweight, strong, and cost-effective composites.

Abstract

In this paper Knowledge Discovery System (KDS) is proposed and implemented for the extraction of knowledge-mean stiffness of a polymer composite material in which when fibers are placed at different orientations. Cosine amplitude method is implemented for retrieving compatible polymer matrix and reinforcement fiber which is coming under predicted fiber class, from the polymer and reinforcement database respectively, based on the design requirements. Fuzzy classification rules to classify fibers into short, medium and long fiber classes are derived based on the fiber length and the computed or derive critical length of fiber. Longitudinal and Transverse module of Polymer Matrix Composite consisting of seven layers with different fiber volume fractions and different fibers orientations at 0,15,30,45,60,75 and 90 degrees are analyzed through Rule-of Mixture material design model. The analysis results are represented in different graphical steps and have been measured with statistical parameters. This data mining application implemented here has focused the mechanical problems of material design and analysis. Therefore, this system is an expert decision support system for optimizing the materials performance for designing light-weight and strong, and cost effective polymer composite materials.