Authors: Abhendra Pratap Singh, Nandini Sharma, Vanshika Dua, Arpit Dwivedi, Aakriti Sharma
Abstract: Composite materials are generated by intermingling two or more diverse components that are individually not able to do various tasks but when put together have become critically important in modern engineering due to their superior mechanical and structural traits. Fiber reinforced polymer (FRP) composites are utilized frequently in the aerospace automotive and construction industries more prominently. Despite their growing adoption, a continuing dilemma involves assessing natural fiber reinforced polymers (NFRP) over synthetic fiber reinforced polymers (SFRP) which differ greatly at the levels of performance cost and environmental impact. Both natural and synthetic composites have their own benefits and drawbacks such that synthetic composites offer excellent strength and durability and natural composites are gaining popularity due to their lightweight renewability and sustainability. This lack of unambiguous data driven comparison often leads to unclear judgment and leads to confusion in choosing the most viable composite for certain technical objectives. To eradicate this gap, the study examines three natural composites flax FRP, hemp FRP and jute FRP and three synthetic composites glass FRP, carbon FRP and aramid FRP. The paper uses computationally intensive analysis and machine learning methods such as linear regression and support vector machine (SVM) to figure out four crucial properties which mostly defines about the composite materials namely density, tensile strength, elastic modulus and moisture absorption. The visualized results of matplotlib based graphs provide a clear insight of how natural and synthetic composites perform individually and collectively through comparative analysis. This research incorporates AI assisted analytical modeling with scientific visualization to give a systematic and sustainable structure for selecting innovative composite materials.
DOI: http://doi.org/10.5281/zenodo.17481163
