Saved in:
Bibliographic Details
Main Authors: Kazemi-Khasragh, Elaheh, Gonzaleza, Carlos, Haranczyk, Maciej
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2401.09139
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917569695318016
author Kazemi-Khasragh, Elaheh
Gonzaleza, Carlos
Haranczyk, Maciej
author_facet Kazemi-Khasragh, Elaheh
Gonzaleza, Carlos
Haranczyk, Maciej
contents The prediction of mechanical and thermal properties of polymers is a critical aspect for polymer development. Herein, we discuss the use of transfer learning approach to predict multiple properties of linear polymers. The neural network model is initially trained to predict the heat capacity in constant pressure (Cp) of linear polymers. Once, the pre-trained model is transferred to predict four additional properties of polymers: specific heat capacity (Cv), shear modulus, flexural stress strength at yield, and tensile creep compliance. They represent a diverse set of mechanical, thermal, and rheological properties. We demonstrate the effectiveness of the approach by achieving high accuracy in predicting the four additional properties using relatively small datasets of 13 to 18 samples. Also, the performance of the base model is examined using five different loss functions. Our results suggest that the combined loss function had better performance compared to the individual loss functions.
format Preprint
id arxiv_https___arxiv_org_abs_2401_09139
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Toward Diverse Polymer Property Prediction Using Transfer Learning
Kazemi-Khasragh, Elaheh
Gonzaleza, Carlos
Haranczyk, Maciej
Soft Condensed Matter
The prediction of mechanical and thermal properties of polymers is a critical aspect for polymer development. Herein, we discuss the use of transfer learning approach to predict multiple properties of linear polymers. The neural network model is initially trained to predict the heat capacity in constant pressure (Cp) of linear polymers. Once, the pre-trained model is transferred to predict four additional properties of polymers: specific heat capacity (Cv), shear modulus, flexural stress strength at yield, and tensile creep compliance. They represent a diverse set of mechanical, thermal, and rheological properties. We demonstrate the effectiveness of the approach by achieving high accuracy in predicting the four additional properties using relatively small datasets of 13 to 18 samples. Also, the performance of the base model is examined using five different loss functions. Our results suggest that the combined loss function had better performance compared to the individual loss functions.
title Toward Diverse Polymer Property Prediction Using Transfer Learning
topic Soft Condensed Matter
url https://arxiv.org/abs/2401.09139