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...

Full description

Saved in:
Bibliographic Details
Main Authors: Ahmed Ghazi Abdulameer, Muhannad M. Mrah, Maryam Bazerkan, Luttfi A. Al-Haddad, Mustafa I. Al-Karkhi
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