Machine learning-driven power prediction in continuous extrusion of pure titanium for enhanced structural resilience under extreme loading
Abstract The increasing demand for advanced materials capable of withstanding extreme loading conditions, such as those encountered during impact or blast events, underscores the need for innovative approaches in material processing. This study focuses on leveraging machine learning (ML) to enhance...
Saved in:
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Springer
2025-01-01
|
Series: | Discover Materials |
Subjects: | |
Online Access: | https://doi.org/10.1007/s43939-024-00175-6 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
_version_ | 1841544213548761088 |
---|---|
author | Ahmed Ghazi Abdulameer Muhannad M. Mrah Maryam Bazerkan Luttfi A. Al-Haddad Mustafa I. Al-Karkhi |
author_facet | Ahmed Ghazi Abdulameer Muhannad M. Mrah Maryam Bazerkan Luttfi A. Al-Haddad Mustafa I. Al-Karkhi |
author_sort | Ahmed Ghazi Abdulameer |
collection | DOAJ |
description | Abstract The increasing demand for advanced materials capable of withstanding extreme loading conditions, such as those encountered during impact or blast events, underscores the need for innovative approaches in material processing. This study focuses on leveraging machine learning (ML) to enhance predictive accuracy in the continuous extrusion of CP-Titanium Grade 2, a material vital for structural resilience in critical applications. Specifically, an Artificial Neural Network (ANN) model optimized using Stochastic Gradient Descent (SGD) was introduced to forecast power requirements with high precision. The analysis utilized a published dataset that comprises theoretical, numerical, and experimental power calculations as a robust foundation for validation and comparison. A visualization highlighted the influence of process parameters, such as feedstock temperature and extrusion wheel velocity, on structural performance to align with the thematic focus of resilient material design. The ANN-SGD model achieved an RMSE of 0.9954 and a CVRMSE of 11.53% which demonstrated significant improvements in prediction accuracy compared to traditional approaches. By achieving superior alignment with experimental results, the model validated its efficacy as a reliable and efficient tool for understanding and optimizing complex manufacturing processes. This research emphasizes the potential of ML to revolutionize material processing for extreme conditions and contribute to the broader goals of structural resilience and sustainable manufacturing. |
format | Article |
id | doaj-art-c75c6ba150ee44309069ec58ccc023d4 |
institution | Kabale University |
issn | 2730-7727 |
language | English |
publishDate | 2025-01-01 |
publisher | Springer |
record_format | Article |
series | Discover Materials |
spelling | doaj-art-c75c6ba150ee44309069ec58ccc023d42025-01-12T12:44:08ZengSpringerDiscover Materials2730-77272025-01-015111710.1007/s43939-024-00175-6Machine learning-driven power prediction in continuous extrusion of pure titanium for enhanced structural resilience under extreme loadingAhmed Ghazi Abdulameer0Muhannad M. Mrah1Maryam Bazerkan2Luttfi A. Al-Haddad3Mustafa I. Al-Karkhi4Training and Workshops Center, University of Technology- IraqMechanical Engineering Department, University of Technology- IraqTraining and Workshops Center, University of Technology- IraqTraining and Workshops Center, University of Technology- IraqMechanical Engineering Department, University of Technology- IraqAbstract The increasing demand for advanced materials capable of withstanding extreme loading conditions, such as those encountered during impact or blast events, underscores the need for innovative approaches in material processing. This study focuses on leveraging machine learning (ML) to enhance predictive accuracy in the continuous extrusion of CP-Titanium Grade 2, a material vital for structural resilience in critical applications. Specifically, an Artificial Neural Network (ANN) model optimized using Stochastic Gradient Descent (SGD) was introduced to forecast power requirements with high precision. The analysis utilized a published dataset that comprises theoretical, numerical, and experimental power calculations as a robust foundation for validation and comparison. A visualization highlighted the influence of process parameters, such as feedstock temperature and extrusion wheel velocity, on structural performance to align with the thematic focus of resilient material design. The ANN-SGD model achieved an RMSE of 0.9954 and a CVRMSE of 11.53% which demonstrated significant improvements in prediction accuracy compared to traditional approaches. By achieving superior alignment with experimental results, the model validated its efficacy as a reliable and efficient tool for understanding and optimizing complex manufacturing processes. This research emphasizes the potential of ML to revolutionize material processing for extreme conditions and contribute to the broader goals of structural resilience and sustainable manufacturing.https://doi.org/10.1007/s43939-024-00175-6Machine learningMaterialsDirect extrusionStructural resilienceExtreme loading |
spellingShingle | Ahmed Ghazi Abdulameer Muhannad M. Mrah Maryam Bazerkan Luttfi A. Al-Haddad Mustafa I. Al-Karkhi Machine learning-driven power prediction in continuous extrusion of pure titanium for enhanced structural resilience under extreme loading Discover Materials Machine learning Materials Direct extrusion Structural resilience Extreme loading |
title | Machine learning-driven power prediction in continuous extrusion of pure titanium for enhanced structural resilience under extreme loading |
title_full | Machine learning-driven power prediction in continuous extrusion of pure titanium for enhanced structural resilience under extreme loading |
title_fullStr | Machine learning-driven power prediction in continuous extrusion of pure titanium for enhanced structural resilience under extreme loading |
title_full_unstemmed | Machine learning-driven power prediction in continuous extrusion of pure titanium for enhanced structural resilience under extreme loading |
title_short | Machine learning-driven power prediction in continuous extrusion of pure titanium for enhanced structural resilience under extreme loading |
title_sort | machine learning driven power prediction in continuous extrusion of pure titanium for enhanced structural resilience under extreme loading |
topic | Machine learning Materials Direct extrusion Structural resilience Extreme loading |
url | https://doi.org/10.1007/s43939-024-00175-6 |
work_keys_str_mv | AT ahmedghaziabdulameer machinelearningdrivenpowerpredictionincontinuousextrusionofpuretitaniumforenhancedstructuralresilienceunderextremeloading AT muhannadmmrah machinelearningdrivenpowerpredictionincontinuousextrusionofpuretitaniumforenhancedstructuralresilienceunderextremeloading AT maryambazerkan machinelearningdrivenpowerpredictionincontinuousextrusionofpuretitaniumforenhancedstructuralresilienceunderextremeloading AT luttfiaalhaddad machinelearningdrivenpowerpredictionincontinuousextrusionofpuretitaniumforenhancedstructuralresilienceunderextremeloading AT mustafaialkarkhi machinelearningdrivenpowerpredictionincontinuousextrusionofpuretitaniumforenhancedstructuralresilienceunderextremeloading |