Machine learning-based study of hardness in polypropylene/carbon nanotube and low-density polyethylene/carbon nanotube composites
Abstract In the present work, the hardness prediction of polypropylene/carbon nanotubes (PP/CNT) and low-density polyethylene/carbon nanotubes (LDPE/CNT) composite materials, processed by microwave technique, has been explored using machine learning models i.e. (Random Forest, Support Vector Regress...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2025-01-01
|
| Series: | Discover Materials |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s43939-025-00176-z |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841559017864822784 |
|---|---|
| author | Harshit Sharma Gaurav Arora Raj Kumar Suman Debnath Suchart Siengchin |
| author_facet | Harshit Sharma Gaurav Arora Raj Kumar Suman Debnath Suchart Siengchin |
| author_sort | Harshit Sharma |
| collection | DOAJ |
| description | Abstract In the present work, the hardness prediction of polypropylene/carbon nanotubes (PP/CNT) and low-density polyethylene/carbon nanotubes (LDPE/CNT) composite materials, processed by microwave technique, has been explored using machine learning models i.e. (Random Forest, Support Vector Regression, K-Nearest Neighbors, Linear Regression, and Neural Network). Four input vectors have been used in the construction of proposed network, such as CNT concentration, power, pressure applied, and exposure time. Hardness prediction is one output that has been evolved from the proposed work. This study presents the prediction of hardness based on machine learning models for both PP/CNT and LDPE/CNT composite materials, and the results show that the Random Forest model consistently performs better than the others models in context with performance metrics like Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Rate of determination (R2) values. Investigations have been performed on resampling strategies, showing that the jackknife approach enhances model precision and robustness in the case of LDPE/CNT composites. For PP/CNT composite material, it has been noticed that Random Forest gives the highest value of R2 (0.94), whereas Random Forest has the lowest R2 value 0.18 for LDPE/CNT composite material. Random Forest is the most reliable model for predicting the characteristics of PP/CNT composite material due to its ability to handle complex datasets. The LDPE/CNT composite material demonstrates superior prediction accuracy, with a maximum error of just 1.61%, making it a better option for high-precision applications due to enhanced mechanical interactions and improved CNT dispersion. |
| format | Article |
| id | doaj-art-91fb544678544a19a9a0ac5ae4239552 |
| institution | Kabale University |
| issn | 2730-7727 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Springer |
| record_format | Article |
| series | Discover Materials |
| spelling | doaj-art-91fb544678544a19a9a0ac5ae42395522025-01-05T12:49:42ZengSpringerDiscover Materials2730-77272025-01-015111810.1007/s43939-025-00176-zMachine learning-based study of hardness in polypropylene/carbon nanotube and low-density polyethylene/carbon nanotube compositesHarshit Sharma0Gaurav Arora1Raj Kumar2Suman Debnath3Suchart Siengchin4Department of Mechanical Engineering, Chandigarh UniversityNatural Composites Research Group Lab, Department of Materials and Production Engineering, The Sirindhorn International Thai-German School of Engineering (TGGS), King Mongkut’s University of Technology North Bangkok (KMUTNB)Department of Mechanical Engineering, Chandigarh UniversityDepartment of Mechanical Engineering, Chandigarh UniversityNatural Composites Research Group Lab, Department of Materials and Production Engineering, The Sirindhorn International Thai-German School of Engineering (TGGS), King Mongkut’s University of Technology North Bangkok (KMUTNB)Abstract In the present work, the hardness prediction of polypropylene/carbon nanotubes (PP/CNT) and low-density polyethylene/carbon nanotubes (LDPE/CNT) composite materials, processed by microwave technique, has been explored using machine learning models i.e. (Random Forest, Support Vector Regression, K-Nearest Neighbors, Linear Regression, and Neural Network). Four input vectors have been used in the construction of proposed network, such as CNT concentration, power, pressure applied, and exposure time. Hardness prediction is one output that has been evolved from the proposed work. This study presents the prediction of hardness based on machine learning models for both PP/CNT and LDPE/CNT composite materials, and the results show that the Random Forest model consistently performs better than the others models in context with performance metrics like Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Rate of determination (R2) values. Investigations have been performed on resampling strategies, showing that the jackknife approach enhances model precision and robustness in the case of LDPE/CNT composites. For PP/CNT composite material, it has been noticed that Random Forest gives the highest value of R2 (0.94), whereas Random Forest has the lowest R2 value 0.18 for LDPE/CNT composite material. Random Forest is the most reliable model for predicting the characteristics of PP/CNT composite material due to its ability to handle complex datasets. The LDPE/CNT composite material demonstrates superior prediction accuracy, with a maximum error of just 1.61%, making it a better option for high-precision applications due to enhanced mechanical interactions and improved CNT dispersion.https://doi.org/10.1007/s43939-025-00176-zPolypropylene/carbon nanotubes (PP/CNT)Low-density polyethylene/carbon nanotubes (LDPE/CNT)Random ForestSupport Vector RegressionK-Nearest NeighborLinear Regression |
| spellingShingle | Harshit Sharma Gaurav Arora Raj Kumar Suman Debnath Suchart Siengchin Machine learning-based study of hardness in polypropylene/carbon nanotube and low-density polyethylene/carbon nanotube composites Discover Materials Polypropylene/carbon nanotubes (PP/CNT) Low-density polyethylene/carbon nanotubes (LDPE/CNT) Random Forest Support Vector Regression K-Nearest Neighbor Linear Regression |
| title | Machine learning-based study of hardness in polypropylene/carbon nanotube and low-density polyethylene/carbon nanotube composites |
| title_full | Machine learning-based study of hardness in polypropylene/carbon nanotube and low-density polyethylene/carbon nanotube composites |
| title_fullStr | Machine learning-based study of hardness in polypropylene/carbon nanotube and low-density polyethylene/carbon nanotube composites |
| title_full_unstemmed | Machine learning-based study of hardness in polypropylene/carbon nanotube and low-density polyethylene/carbon nanotube composites |
| title_short | Machine learning-based study of hardness in polypropylene/carbon nanotube and low-density polyethylene/carbon nanotube composites |
| title_sort | machine learning based study of hardness in polypropylene carbon nanotube and low density polyethylene carbon nanotube composites |
| topic | Polypropylene/carbon nanotubes (PP/CNT) Low-density polyethylene/carbon nanotubes (LDPE/CNT) Random Forest Support Vector Regression K-Nearest Neighbor Linear Regression |
| url | https://doi.org/10.1007/s43939-025-00176-z |
| work_keys_str_mv | AT harshitsharma machinelearningbasedstudyofhardnessinpolypropylenecarbonnanotubeandlowdensitypolyethylenecarbonnanotubecomposites AT gauravarora machinelearningbasedstudyofhardnessinpolypropylenecarbonnanotubeandlowdensitypolyethylenecarbonnanotubecomposites AT rajkumar machinelearningbasedstudyofhardnessinpolypropylenecarbonnanotubeandlowdensitypolyethylenecarbonnanotubecomposites AT sumandebnath machinelearningbasedstudyofhardnessinpolypropylenecarbonnanotubeandlowdensitypolyethylenecarbonnanotubecomposites AT suchartsiengchin machinelearningbasedstudyofhardnessinpolypropylenecarbonnanotubeandlowdensitypolyethylenecarbonnanotubecomposites |