Exploration of the Application of Data-Driven and Generation Models in the Design of Thermoplastic Vulcanizate Rubbers

TL;DR

This paper explores using data-driven and generation models to design thermoplastic vulcanizate rubbers, improving predictions and material development.

Contribution

The study introduces innovative data-driven and generation models for predicting and designing TPV rubber properties.

Findings

Predicted values for TPV rubber properties align well with experimental data.

Generation models successfully produce new material formula data for TPV rubber.

The approach could reduce R&D costs and speed up new material development.

Abstract

The rapid advancement of big data and artificial intelligence has highlighted the substantial potential of data-driven approaches in polymer material research and development. In the present study, data-driven predictive models were developed to accurately forecast the density, tensile strength, flexural strength and melt mass flow rate of thermoplastic vulcanizate (TPV) rubber. Furthermore, a generation model was used to produce new material formula data for TPV rubber, and predictions were made for the aforementioned properties. The results indicated that the predicted values are in good agreement with experimental data. This study introduces innovative strategies and methodologies for the intelligent design of polymer materials, which could potentially lower research and development costs and accelerate the emergence of novel materials.

Figures