Dataset on graphite nanoplatelet enhanced HDPE composites: An ensemble machine learning approach estimating the tensile modulus, toughness, and hardness thus creating a roadmap for new product developmentMendeley Data

Abstract

This article presents a dataset examining the impact of Graphite Nanoplatelets (GNP), injection pressure, and injection temperature on the mechanical properties of High-Density Polyethylene (HDPE) composites. The study uses an ensemble machine learning approach to build a predictive model for response variables: tensile modulus, toughness, and hardness. The model used was a Random Forest regressor, which was robust to outliers and data distribution types. Grid Search CV and Random Search CV were used for hyperparameter tuning, and predictive analysis of model parameters was conducted using both. The response variables were divided into high, medium, and low classes based on the quartile distributions. A Decision Tree classifier was used to derive rules for each class. These methods and the comprehensive dataset and analysis offer a robust foundation for future research in polymer composites. The insights can be leveraged with further research to develop the highest grade of polymer composite products. This might include new applications, improved performance, or cost reduction by optimizing the resources. This research may help in developing a detailed concept for the viable ideas and testing of a new polymer-based product thus bringing a significant impact in the polymer industry as a whole.

Keywords

• Mechanical properties
• Machine learning
• Random forest
• Polymer composite
• Graphite