Hybrid Polymer Composite Tensile Strength Estimation using K-Nearest Neighboring Classification Algorithm

Articles in Press

Document Type : Research Article

Authors

1 Department of Mechanical Engineering, School of Engineering and Applied Sciences, Bennett University, Greater Noida, UP, India

2 R.H. SAPAT College of Engineering, Management Studies and Research, Maharashtra, India

3 University Centre for Research and Development, Chandigarh University, Punjab, India

4 Associate Professor, Department of Aeronautical Engineering, Parul Institute of Engineering and Technology, Parul University, Vadodara, India

Abstract

The aim of this research work is to characterize the tensile strength of ABS-Cu and ABS-Al composites of different proportions of percentage compositions, as well as the incorporation of surfactant material. For the analysis carried out in the present study, the k-Nearest Neighboring (kNN) classification algorithm is used in order to predict the tensile strength of the various compositions of the ABS-Al and ABS-Cu composites. Real data was not used to train the model due to the time-consuming process; instead, they resorted to synthetic data for the classification model, and for the tensile strength data, they were trained and predicted with better results. The kNN classification algorithm of the ABS-Cu predicted the k-value accuracy to be 80% for k=1 and k=2, and 85% for k=3 and k=5. Similarly, the prediction accuracy for the ABS-Al composition yielded the same results: As the value of k is increased, the required percentage of samples is 80% for k=1 and k=2, 85% for k=3, and 90% for k=5, respectively. The kNN classification algorithm model was also successful in predicting tensile strength, with a recall of more than 80% and an F1 score of 90-95%. A higher quantity of copper and aluminium is said to have the ability to improve the tensile strength of the specimens.

Keywords

Main Subjects

Mechanics of Advanced Composite Structures

Articles in Press, Accepted Manuscript
Available Online from 17 July 2025

Files

Share

How to cite

Statistics